-------------------------------------------------------------------------------------------------
      name:  <unnamed>
       log:  /Users/Wei/Dropbox/Fertility/Results/Robust.log
  log type:  text
 opened on:  21 Apr 2020, 00:02:30

. 
. * Figure c4 
. 
. **************************
. use  "$path2/marr_policy", clear

. gen early_marr = 1 - late_marr

. replace high_occ = 0 if work == 0
(1,603,442 real changes made)

. su senior college married_ever early_marr work high_occ [aw = wt] if men, sep(100)

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
      senior | 5878746     8071776    .2230751   .4163083          0          1
     college | 5878746     8071776    .0698561   .2549044          0          1
married_ever | 5878746     8071776    .9155201   .2781062          0          1
  early_marr | 5878746     8071776    .7232607   .4473865          0          1
        work | 5878746     8071776    .9067503   .2907821          0          1
    high_occ | 5878746     8071776    .1276292   .3336765          0          1

. su senior college married_ever early_marr work high_occ [aw = wt] if women, sep(100)

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
      senior | 5668820     7871897    .1577345   .3644919          0          1
     college | 5668820     7871897    .0460877   .2096752          0          1
married_ever | 5668820     7871897     .979014   .1433375          0          1
  early_marr | 5668820     7871897    .8935499   .3084128          0          1
        work | 5668820     7871897    .7676171   .4223519          0          1
    high_occ | 5668820     7871897    .0830444   .2759494          0          1

. su senior college married_ever early_marr work high_occ [aw = wt] if men & age <= 30, sep(100)

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
      senior | 1371132     1682448    .2691637   .4435254          0          1
     college | 1371132     1682448    .0869542   .2817681          0          1
married_ever | 1371132     1682448    .7792259   .4147687          0          1
  early_marr | 1371132     1682448    .6930966   .4612091          0          1
        work | 1371132     1682448    .9419982   .2337469          0          1
    high_occ | 1371132     1682448    .1192084   .3240337          0          1

. su senior college married_ever early_marr work high_occ [aw = wt] if men & age > 30, sep(100)

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
      senior | 4507614     6389328    .2109389   .4079752          0          1
     college | 4507614     6389328    .0653538   .2471492          0          1
married_ever | 4507614     6389328    .9514093   .2150108          0          1
  early_marr | 4507614     6389328    .7312035   .4433339          0          1
        work | 4507614     6389328    .8974687   .3033457          0          1
    high_occ | 4507614     6389328    .1298465   .3361345          0          1

. su senior college married_ever early_marr work high_occ [aw = wt] if women & age <= 30, sep(100
> )

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
      senior | 1353760     1690618    .2190968   .4136345          0          1
     college | 1353760     1690618    .0718512   .2582415          0          1
married_ever | 1353760     1690618    .9223314   .2676495          0          1
  early_marr | 1353760     1690618    .8696175   .3367239          0          1
        work | 1353760     1690618    .8275051   .3778102          0          1
    high_occ | 1353760     1690618    .1043388   .3056996          0          1

. su senior college married_ever early_marr work high_occ [aw = wt] if women & age > 30, sep(100)

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
      senior | 4315060     6181279    .1409516   .3479716          0          1
     college | 4315060     6181279    .0390413   .1936932          0          1
married_ever | 4315060     6181279     .994517   .0738441          0          1
  early_marr | 4315060     6181279    .9000956   .2998725          0          1
        work | 4315060     6181279    .7512374    .432296          0          1
    high_occ | 4315060     6181279    .0772203   .2669407          0          1

. 
. use "$path2/fines_economic", clear 

. drop lngdp_pc lngdp secondary primary

. merge 1:1 prov year using "$path2/60years", keepusing(gdp_pc teacher_primary teacher_second)

    Result                           # of obs.
    -----------------------------------------
    not matched                             1
        from master                         1  (_merge==1)
        from using                          0  (_merge==2)

    matched                             1,859  (_merge==3)
    -----------------------------------------

. ren teacher_primary primary 

. ren teacher_second secondary

. gen lngdp_pc = ln(gdp_pc)
(84 missing values generated)

. save "$path2/fines_economic_new",replace 
file /Users/Wei/Dropbox/Fertility/Workingdata/fines_economic_new.dta saved

. 
. use "$path2/marr_policy", clear

. collapse men, by(prov year_birth)

. gen sex_ratio = men/(1-men)

. ren men male_prop

. save "$path2/sex_ratio_prov_year", replace 
file /Users/Wei/Dropbox/Fertility/Workingdata/sex_ratio_prov_year.dta saved

. 
. 
. 
. set more off

. 
. use "$path2/marr_policy", clear

. gen mort_rate = 100 -sur_rate if women == 1 
(6,645,526 missing values generated)

. ren year yr_t

. gen year = year_birth +15

. merge m:1 prov year using "$path2/fines_economic_new", keepusing(lngdp_pc secondary primary)
(note: variable prov was byte, now float to accommodate using data's values)

    Result                           # of obs.
    -----------------------------------------
    not matched                           589
        from master                         0  (_merge==1)
        from using                        589  (_merge==2)

    matched                        11,547,566  (_merge==3)
    -----------------------------------------

. drop if _merge == 2 
(589 observations deleted)

. drop _merge 

. drop year

. ren yr_t year

. egen male_prop = mean(men), by(prov year_birth)

. gen sex_ratio = male_prop/(1-male_prop)

. 
. 
. egen fine_6_20 = rowmean(fine_age6-fine_age20)

. drop fine_age*

. gl FINE_ALL = "fine_6_15 fine_16_20 fine_21_25 fine_26_30"

. gl FINE_VAR_4 =  "fine_6_15 fine_16_20 fine_21_30"

. 
. gl FINE_VAR_3 =  "fine_6_15 fine_16_20 fine_21_25"

. gl FINE_VAR_1 = "fine_6_20"

. 
. replace nbroth = 99 if mi(nbroth)
(11,325,299 real changes made)

. replace nsister = 99 if mi(nsister)
(11,325,299 real changes made)

. 
. gen lnsec = ln(secondary)
(577,564 missing values generated)

. gen lnpri = ln(primary)
(352,981 missing values generated)

. gen lnsec_mi = mi(lnsec)

. gen lnpri_mi = mi(lnpri)

. gen lngdp_pc_mi = mi(lnpri)

. 
. egen mean_lnsec = mean(lnsec)

. egen mean_lnpri = mean(lnpri)

. egen mean_lngdp = mean(lngdp_pc)

. 
. replace lnsec = mean_lnsec if mi(lnsec)
(577,564 real changes made)

. replace lnpri = mean_lnpri if mi(lnpri)
(352,981 real changes made)

. replace lngdp_pc = mean_lngdp if mi(lngdp_pc)
(41,708 real changes made)

. drop mean_lnsec  mean_lnpri mean_lngdp

. 
. gen provid = prov 

. merge m:1 provid using "$path2/Comp_Law", keepusing(law_*) nogen 

    Result                           # of obs.
    -----------------------------------------
    not matched                             1
        from master                         0  
        from using                          1  

    matched                        11,547,566  
    -----------------------------------------

. 
. cap drop elig

. gen elig = 1 if year_birth >= law_year - 6 
(11,539,033 missing values generated)

. replace elig = 0 if year_birth <= law_year - 16
(10,501,045 real changes made)

. replace elig = (16-(law_year - year_birth ))/10 if law_year - 16<year_birth & year_birth<law_ye
> ar - 6 
(1,037,988 real changes made)

. drop provid

. 
. 
. gl CONTROL_1 = "women#prov##c.year_birth women#year#year_birth"

. gl CONTROL_2 = "women#prov##c.year_birth women#year#year_birth women#prov#year"

. 
. gen urban_hk = hktype == 2 

. gen urban_res = city_flag <= 2 if !mi(city_flag)
(4,140,909 missing values generated)

. replace urban_res = urban_hk if  mi(urban_res)
(4,140,909 real changes made)

. set more off 

. replace high_occ = 0 if high_occ == . & !mi(work)
(1,603,442 real changes made)

. 
. egen han_p = mean(han), by(year hhid prov region)
(1 missing value generated)

. replace late_marr = 1- late_marr // early married 
(11,547,566 real changes made)

. 
. set more off

. gen nsib = nbroth + nsister
(1 missing value generated)

. cap erase "$path3/OCP_women_census_control.xls"

. cap erase "$path3/OCP_women_census_control.txt"

. foreach y_var in "senior" "college ""married_ever""late_marr" "work" "high_occ"  {
  2. local j = 1 
  3. gen coef_men_`y_var' = .
  4. gen coef_women_`y_var' = .
  5. gen se_men_`y_var' = .
  6. gen se_women_`y_var' = .
  7. gen mean_`y_var'  = . 
  8. gen std_`y_var'  = . 
  9. su `y_var'  [aw = wt] if han 
 10. replace mean_`y_var' = `r(mean)'  
 11. replace std_`y_var' = `r(sd)'  
 12. local j = 1 
 13. foreach x_var in  "$FINE_VAR_1"{
 14. reghdfe `y_var' women#c.(`x_var') if han [aw = wt], a($CONTROL_1) cluster(prov)
 15. replace coef_men_`y_var' = _b[0.women#`x_var']  in `j'
 16. replace coef_women_`y_var' = _b[1.women#`x_var']   in `j'
 17. replace se_men_`y_var' = _se[0.women#`x_var']  in `j'
 18. replace se_women_`y_var' = _se[1.women#`x_var']   in `j'
 19. outreg2 using "$path3/OCP_women_census_control.xls", dec(3) append
 20. local ++j 
 21. 
. reghdfe `y_var' women#c.(`x_var') if han  [aw = wt], a($CONTROL_1 women#prov#year) cluster(prov
> )
 22. replace coef_men_`y_var' = _b[0.women#`x_var']  in `j'
 23. replace coef_women_`y_var' = _b[1.women#`x_var'] in `j'
 24. replace se_men_`y_var' = _se[0.women#`x_var']  in `j'
 25. replace se_women_`y_var' = _se[1.women#`x_var']   in `j'
 26. outreg2 using "$path3/OCP_women_census_control.xls", dec(3) append
 27. local ++j 
 28. 
. reghdfe `y_var' women#c.(`x_var') if han  [aw = wt], a(women#nsib $CONTROL_1) cluster(prov)
 29. replace coef_men_`y_var' = _b[0.women#`x_var'] in `j'
 30. replace coef_women_`y_var' = _b[1.women#`x_var']  in `j'
 31. replace se_men_`y_var' = _se[0.women#`x_var'] in `j'
 32. replace se_women_`y_var' = _se[1.women#`x_var']  in `j'
 33. outreg2 using "$path3/OCP_women_census_control.xls", dec(3) append
 34. local ++j
 35. 
. reghdfe `y_var' women#c.(`x_var' sex_ratio) if han  [aw = wt], a($CONTROL_1) cluster(prov)
 36. replace coef_men_`y_var' = _b[0.women#`x_var'] in `j'
 37. replace coef_women_`y_var' = _b[1.women#`x_var']  in `j'
 38. replace se_men_`y_var' = _se[0.women#`x_var']  in `j'
 39. replace se_women_`y_var' = _se[1.women#`x_var']  in `j'
 40. outreg2 using "$path3/OCP_women_census_control.xls", dec(3) append
 41. local ++j
 42. 
. reghdfe `y_var' women#c.(`x_var' lngdp_pc) if han  [aw = wt], a(women#lngdp_pc_mi $CONTROL_1) c
> luster(prov)
 43. replace coef_men_`y_var' = _b[0.women#`x_var']  in `j'
 44. replace coef_women_`y_var' = _b[1.women#`x_var']   in `j'
 45. replace se_men_`y_var' = _se[0.women#`x_var']  in `j'
 46. replace se_women_`y_var' = _se[1.women#`x_var']   in `j'
 47. outreg2 using "$path3/OCP_women_census_control.xls", dec(3) append
 48. local ++j
 49. 
. reghdfe `y_var' women#c.(`x_var' lnsec lnpri) if han  [aw = wt], a(women#lnsec_mi  women#lnpri_
> mi women#nsib $CONTROL_1) cluster(prov)
 50. replace coef_men_`y_var' = _b[0.women#`x_var']  in `j'
 51. replace coef_women_`y_var' = _b[1.women#`x_var']  in `j'
 52. replace se_men_`y_var' = _se[0.women#`x_var']  in `j'
 53. replace se_women_`y_var' = _se[1.women#`x_var']  in `j'
 54. outreg2 using "$path3/OCP_women_census_control.xls", dec(3) append
 55. local ++j
 56. 
. reghdfe `y_var' women#c.(`x_var' elig) if han  [aw = wt], a( $CONTROL_1) cluster(prov)
 57. replace coef_men_`y_var' = _b[0.women#`x_var'] in `j'
 58. replace coef_women_`y_var' = _b[1.women#`x_var']  in `j'
 59. replace se_men_`y_var' = _se[0.women#`x_var']  in `j'
 60. replace se_women_`y_var' = _se[1.women#`x_var']  in `j'
 61. outreg2 using "$path3/OCP_women_census_control.xls", dec(3) append
 62. local ++j
 63. 
. 
. reghdfe `y_var' women#c.(`x_var' sex_ratio lngdp_pc lnsec lnpri elig) if han [aw = wt], a(women
> #lngdp_pc_mi women#lnsec_mi  women#lnpri_mi  women#nsib $CONTROL_1) cluster(prov)
 64. replace coef_men_`y_var' = _b[0.women#`x_var']  in `j'
 65. replace coef_women_`y_var' = _b[1.women#`x_var']   in `j'
 66. replace se_men_`y_var' = _se[0.women#`x_var']  in `j'
 67. replace se_women_`y_var' = _se[1.women#`x_var']    in `j'
 68. outreg2 using "$path3/OCP_women_census_control.xls", dec(3) append
 69. local ++j
 70. 
. reghdfe `y_var' women#c.(`x_var' sex_ratio lngdp_pc lnsec lnpri elig) if han [aw = wt], a(women
> #lngdp_pc_mi women#lnsec_mi  women#lnpri_mi  women#nsib $CONTROL_1 women#prov#year) cluster(pro
> v)
 71. replace coef_men_`y_var' = _b[0.women#`x_var']   in `j'
 72. replace coef_women_`y_var' = _b[1.women#`x_var']   in `j'
 73. replace se_men_`y_var' = _se[0.women#`x_var']  in `j'
 74. replace se_women_`y_var' = _se[1.women#`x_var']   in `j'
 75. outreg2 using "$path3/OCP_women_census_control.xls", dec(3) append
 76. local ++j
 77. 
. }
 78. }
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
      senior |10645223    14607038    .1956363   .3966898          0          1
(11,547,567 real changes made)
(11,547,567 real changes made)
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 2 HDFE groups                           F(   2,     30) =       3.21
Statistics robust to heteroskedasticity           Prob > F        =     0.0543
                                                  R-squared       =     0.0720
                                                  Adj R-squared   =     0.0720
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.3821

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
           senior |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0305929   .0142456     2.15   0.040     .0014995    .0596863
               1  |   .0448009   .0186448     2.40   0.023     .0067231    .0828787
                  |
            _cons |     .18102   .0057705    31.37   0.000     .1692351    .1928049
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 8 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 3 HDFE groups                           F(   2,     30) =       3.09
Statistics robust to heteroskedasticity           Prob > F        =     0.0602
                                                  R-squared       =     0.0725
                                                  Adj R-squared   =     0.0724
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.3821

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
           senior |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0292505   .0139092     2.10   0.044     .0008441    .0576568
               1  |   .0441372   .0186858     2.36   0.025     .0059757    .0822987
                  |
            _cons |   .1814055   .0057292    31.66   0.000     .1697049    .1931061
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
           women#prov#year |       184         184           0    *|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 9 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 3 HDFE groups                           F(   2,     30) =       2.82
Statistics robust to heteroskedasticity           Prob > F        =     0.0752
                                                  R-squared       =     0.0755
                                                  Adj R-squared   =     0.0754
                                                  Within R-sq.    =     0.0002
Number of clusters (prov)    =         31         Root MSE        =     0.3814

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
           senior |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0341283   .0144115     2.37   0.025     .0046961    .0635606
               1  |   .0501228   .0239044     2.10   0.045     .0013034    .0989421
                  |
            _cons |   .1793025   .0071508    25.07   0.000     .1646986    .1939065
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#nsib |        26           0          26     |
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204     |
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 2 HDFE groups                           F(   4,     30) =       8.88
Statistics robust to heteroskedasticity           Prob > F        =     0.0001
                                                  R-squared       =     0.0721
                                                  Adj R-squared   =     0.0720
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.3821

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
           senior |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0302529   .0142057     2.13   0.042     .0012411    .0592648
               1  |   .0437301   .0179046     2.44   0.021     .0071641     .080296
                  |
women#c.sex_ratio |
               0  |   .0239247   .0436297     0.55   0.588     -.065179    .1130284
               1  |   .0790544    .030456     2.60   0.014     .0168549    .1412538
                  |
            _cons |   .1285438   .0333149     3.86   0.001     .0605056     .196582
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 7 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 3 HDFE groups                           F(   4,     30) =       4.26
Statistics robust to heteroskedasticity           Prob > F        =     0.0075
                                                  R-squared       =     0.0722
                                                  Adj R-squared   =     0.0722
                                                  Within R-sq.    =     0.0003
Number of clusters (prov)    =         31         Root MSE        =     0.3821

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
           senior |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0238817   .0146115     1.63   0.113    -.0059589    .0537223
               1  |    .036261   .0161989     2.24   0.033     .0031785    .0693435
                  |
 women#c.lngdp_pc |
               0  |   .0472821   .0176166     2.68   0.012     .0113043      .08326
               1  |   .0617506   .0216678     2.85   0.008      .017499    .1060022
                  |
            _cons |  -.1375022   .1015097    -1.35   0.186    -.3448126    .0698083
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         4           0           4     |
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204     |
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 13 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 5 HDFE groups                           F(   6,     30) =       7.30
Statistics robust to heteroskedasticity           Prob > F        =     0.0001
                                                  R-squared       =     0.0755
                                                  Adj R-squared   =     0.0755
                                                  Within R-sq.    =     0.0002
Number of clusters (prov)    =         31         Root MSE        =     0.3814

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
           senior |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0353212   .0142143     2.48   0.019     .0062917    .0643507
               1  |   .0530199   .0236915     2.24   0.033     .0046355    .1014043
                  |
    women#c.lnsec |
               0  |   .0014664   .0060562     0.24   0.810     -.010902    .0138349
               1  |  -.0003696   .0054501    -0.07   0.946    -.0115003    .0107611
                  |
    women#c.lnpri |
               0  |  -.0107671   .0190346    -0.57   0.576    -.0496409    .0281067
               1  |  -.0283349   .0180981    -1.57   0.128    -.0652963    .0086264
                  |
            _cons |   .4012011   .2177157     1.84   0.075    -.0434338    .8458359
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
            women#lnsec_mi |         4           0           4     |
            women#lnpri_mi |         4           2           2     |
                women#nsib |        26           2          24    ?|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 2 HDFE groups                           F(   4,     30) =       3.99
Statistics robust to heteroskedasticity           Prob > F        =     0.0103
                                                  R-squared       =     0.0721
                                                  Adj R-squared   =     0.0720
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.3821

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
           senior |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0309211   .0135567     2.28   0.030     .0032346    .0586077
               1  |   .0449721   .0186246     2.41   0.022     .0069356    .0830085
                  |
     women#c.elig |
               0  |   .0743129   .0319858     2.32   0.027     .0089892    .1396365
               1  |   .0299058   .0328721     0.91   0.370    -.0372279    .0970396
                  |
            _cons |   .1783712   .0055287    32.26   0.000       .16708    .1896624
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 13 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 6 HDFE groups                           F(  12,     30) =       7.07
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.0758
                                                  Adj R-squared   =     0.0757
                                                  Within R-sq.    =     0.0005
Number of clusters (prov)    =         31         Root MSE        =     0.3814

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
           senior |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |    .026511   .0110541     2.40   0.023     .0039354    .0490866
               1  |   .0417796   .0188689     2.21   0.035     .0032443     .080315
                  |
women#c.sex_ratio |
               0  |   .0175043   .0359853     0.49   0.630    -.0559874     .090996
               1  |   .0707768   .0277322     2.55   0.016       .01414    .1274135
                  |
 women#c.lngdp_pc |
               0  |   .0620782   .0185163     3.35   0.002     .0242628    .0998936
               1  |   .0729384   .0244505     2.98   0.006     .0230038    .1228729
                  |
    women#c.lnsec |
               0  |   .0016528   .0050661     0.33   0.746    -.0086935    .0119991
               1  |  -.0015389    .004424    -0.35   0.730    -.0105739    .0074962
                  |
    women#c.lnpri |
               0  |   -.009523   .0154758    -0.62   0.543    -.0411287    .0220827
               1  |  -.0247405   .0135321    -1.83   0.077    -.0523767    .0028958
                  |
     women#c.elig |
               0  |   .0571047   .0291808     1.96   0.060    -.0024903    .1166998
               1  |   .0120841   .0322611     0.37   0.711    -.0538019    .0779702
                  |
            _cons |  -.0631126   .1905145    -0.33   0.743    -.4521952    .3259699
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         4           0           4     |
            women#lnsec_mi |         4           2           2     |
            women#lnpri_mi |         4           4           0    ?|
                women#nsib |        26           2          24    ?|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 15 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 7 HDFE groups                           F(  12,     30) =       7.13
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.0762
                                                  Adj R-squared   =     0.0762
                                                  Within R-sq.    =     0.0005
Number of clusters (prov)    =         31         Root MSE        =     0.3813

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
           senior |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0252352   .0107146     2.36   0.025     .0033532    .0471173
               1  |   .0410018   .0186145     2.20   0.035     .0029859    .0790177
                  |
women#c.sex_ratio |
               0  |   .0218916   .0358996     0.61   0.547    -.0514252    .0952084
               1  |   .0721881   .0271448     2.66   0.012     .0167509    .1276252
                  |
 women#c.lngdp_pc |
               0  |   .0615321   .0182688     3.37   0.002     .0242222     .098842
               1  |   .0737925   .0247329     2.98   0.006     .0232812    .1243038
                  |
    women#c.lnsec |
               0  |   .0014755   .0046164     0.32   0.751    -.0079524    .0109034
               1  |  -.0015884   .0040619    -0.39   0.699    -.0098838    .0067071
                  |
    women#c.lnpri |
               0  |  -.0105104   .0153127    -0.69   0.498    -.0417831    .0207623
               1  |  -.0256268   .0134493    -1.91   0.066    -.0530939    .0018402
                  |
     women#c.elig |
               0  |    .054471     .03072     1.77   0.086    -.0082676    .1172097
               1  |   .0124775   .0323375     0.39   0.702    -.0535645    .0785196
                  |
            _cons |  -.0542958   .1838578    -0.30   0.770    -.4297834    .3211919
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         4           0           4     |
            women#lnsec_mi |         4           2           2     |
            women#lnpri_mi |         4           4           0    ?|
                women#nsib |        26           2          24    ?|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204    ?|
           women#prov#year |       184         184           0    *|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
     college |10645223    14607038    .0592826   .2361528          0          1
(11,547,567 real changes made)
(11,547,567 real changes made)
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 2 HDFE groups                           F(   2,     30) =       1.92
Statistics robust to heteroskedasticity           Prob > F        =     0.1647
                                                  R-squared       =     0.0334
                                                  Adj R-squared   =     0.0334
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.2322

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
          college |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0176939   .0106179     1.67   0.106    -.0039908    .0393787
               1  |   .0174068    .009149     1.90   0.067     -.001278    .0360915
                  |
            _cons |   .0525003   .0037682    13.93   0.000     .0448046    .0601961
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 8 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 3 HDFE groups                           F(   2,     30) =       1.74
Statistics robust to heteroskedasticity           Prob > F        =     0.1919
                                                  R-squared       =     0.0339
                                                  Adj R-squared   =     0.0338
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.2321

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
          college |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0162651   .0103599     1.57   0.127    -.0048926    .0374229
               1  |   .0162483   .0089849     1.81   0.081    -.0021013    .0345979
                  |
            _cons |   .0529994   .0036865    14.38   0.000     .0454706    .0605283
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
           women#prov#year |       184         184           0    *|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 9 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 3 HDFE groups                           F(   2,     30) =       2.89
Statistics robust to heteroskedasticity           Prob > F        =     0.0710
                                                  R-squared       =     0.0378
                                                  Adj R-squared   =     0.0378
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.2317

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
          college |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0196606   .0085749     2.29   0.029     .0021483     .037173
               1  |   .0207222   .0086448     2.40   0.023     .0030671    .0383772
                  |
            _cons |   .0514752   .0032645    15.77   0.000     .0448081    .0581422
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#nsib |        26           0          26     |
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204     |
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 2 HDFE groups                           F(   4,     30) =       2.25
Statistics robust to heteroskedasticity           Prob > F        =     0.0877
                                                  R-squared       =     0.0334
                                                  Adj R-squared   =     0.0334
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.2322

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
          college |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0177198   .0107854     1.64   0.111    -.0043069    .0397464
               1  |   .0173886   .0092574     1.88   0.070    -.0015175    .0362947
                  |
women#c.sex_ratio |
               0  |  -.0018193   .0182621    -0.10   0.921    -.0391155     .035477
               1  |   .0013431   .0128105     0.10   0.917    -.0248194    .0275056
                  |
            _cons |   .0527688    .012178     4.33   0.000      .027898    .0776395
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 7 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 3 HDFE groups                           F(   4,     30) =       3.06
Statistics robust to heteroskedasticity           Prob > F        =     0.0315
                                                  R-squared       =     0.0335
                                                  Adj R-squared   =     0.0335
                                                  Within R-sq.    =     0.0002
Number of clusters (prov)    =         31         Root MSE        =     0.2322

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
          college |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0213184    .009924     2.15   0.040     .0010509    .0415859
               1  |   .0208539   .0084284     2.47   0.019     .0036408     .038067
                  |
 women#c.lngdp_pc |
               0  |  -.0254329   .0162682    -1.56   0.128    -.0586569    .0077911
               1  |  -.0248815   .0148203    -1.68   0.104    -.0551487    .0053856
                  |
            _cons |   .1996814   .0916302     2.18   0.037     .0125476    .3868153
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         4           0           4     |
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204     |
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 13 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 5 HDFE groups                           F(   6,     30) =       2.27
Statistics robust to heteroskedasticity           Prob > F        =     0.0629
                                                  R-squared       =     0.0378
                                                  Adj R-squared   =     0.0378
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.2317

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
          college |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0190406   .0085215     2.23   0.033     .0016374    .0364437
               1  |   .0203077   .0086704     2.34   0.026     .0026003     .038015
                  |
    women#c.lnsec |
               0  |  -.0041144   .0020385    -2.02   0.053    -.0082775    .0000487
               1  |  -.0016047   .0013725    -1.17   0.252    -.0044077    .0011983
                  |
    women#c.lnpri |
               0  |   .0043228   .0042303     1.02   0.315    -.0043165    .0129622
               1  |   .0029276   .0033421     0.88   0.388     -.003898    .0097531
                  |
            _cons |   .0400051   .0343858     1.16   0.254    -.0302202    .1102303
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
            women#lnsec_mi |         4           0           4     |
            women#lnpri_mi |         4           2           2     |
                women#nsib |        26           2          24    ?|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 2 HDFE groups                           F(   4,     30) =       1.35
Statistics robust to heteroskedasticity           Prob > F        =     0.2754
                                                  R-squared       =     0.0334
                                                  Adj R-squared   =     0.0334
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.2322

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
          college |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0177931   .0105352     1.69   0.102    -.0037225    .0393088
               1  |   .0175015   .0091123     1.92   0.064    -.0011084    .0361113
                  |
     women#c.elig |
               0  |   .0224565   .0178455     1.26   0.218    -.0139888    .0589019
               1  |   .0165505   .0141944     1.17   0.253    -.0124382    .0455393
                  |
            _cons |   .0514969   .0039464    13.05   0.000     .0434372    .0595565
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 13 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 6 HDFE groups                           F(  12,     30) =       3.57
Statistics robust to heteroskedasticity           Prob > F        =     0.0023
                                                  R-squared       =     0.0378
                                                  Adj R-squared   =     0.0378
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.2316

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
          college |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0211612   .0082323     2.57   0.015     .0043485    .0379739
               1  |   .0222296   .0082543     2.69   0.011     .0053721    .0390872
                  |
women#c.sex_ratio |
               0  |    .011917   .0187831     0.63   0.531    -.0264432    .0502773
               1  |   .0154355   .0106229     1.45   0.157    -.0062594    .0371304
                  |
 women#c.lngdp_pc |
               0  |  -.0158647   .0150343    -1.06   0.300    -.0465688    .0148394
               1  |  -.0155238   .0131938    -1.18   0.249    -.0424691    .0114215
                  |
    women#c.lnsec |
               0  |  -.0038924   .0021611    -1.80   0.082    -.0083059    .0005211
               1  |  -.0014988   .0014583    -1.03   0.312    -.0044769    .0014794
                  |
    women#c.lnpri |
               0  |   .0036311   .0052881     0.69   0.498    -.0071687    .0144309
               1  |   .0024625   .0036584     0.67   0.506    -.0050089     .009934
                  |
     women#c.elig |
               0  |  -.0028384   .0161616    -0.18   0.862    -.0358449    .0301681
               1  |  -.0072943   .0128847    -0.57   0.576    -.0336084    .0190198
                  |
            _cons |   .1230933    .109305     1.13   0.269    -.1001372    .3463237
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         4           0           4     |
            women#lnsec_mi |         4           2           2     |
            women#lnpri_mi |         4           4           0    ?|
                women#nsib |        26           2          24    ?|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 15 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 7 HDFE groups                           F(  12,     30) =       3.08
Statistics robust to heteroskedasticity           Prob > F        =     0.0061
                                                  R-squared       =     0.0383
                                                  Adj R-squared   =     0.0382
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.2316

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
          college |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |    .019925   .0079422     2.51   0.018     .0037048    .0361453
               1  |    .021308   .0080027     2.66   0.012     .0049642    .0376517
                  |
women#c.sex_ratio |
               0  |   .0140465   .0188618     0.74   0.462    -.0244743    .0525674
               1  |   .0148535   .0109677     1.35   0.186    -.0075455    .0372525
                  |
 women#c.lngdp_pc |
               0  |  -.0153348   .0146331    -1.05   0.303    -.0452197    .0145501
               1  |   -.015581   .0129876    -1.20   0.240    -.0421052    .0109432
                  |
    women#c.lnsec |
               0  |  -.0033244   .0020897    -1.59   0.122     -.007592    .0009433
               1  |  -.0011622   .0015065    -0.77   0.446    -.0042388    .0019145
                  |
    women#c.lnpri |
               0  |   .0018903   .0051989     0.36   0.719    -.0087272    .0125078
               1  |   .0013482   .0036316     0.37   0.713    -.0060686    .0087649
                  |
     women#c.elig |
               0  |  -.0008689   .0160849    -0.05   0.957    -.0337187    .0319809
               1  |  -.0050239   .0128925    -0.39   0.700     -.031354    .0213062
                  |
            _cons |   .1330978   .1088568     1.22   0.231    -.0892174    .3554131
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         4           0           4     |
            women#lnsec_mi |         4           2           2     |
            women#lnpri_mi |         4           4           0    ?|
                women#nsib |        26           2          24    ?|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204    ?|
           women#prov#year |       184         184           0    *|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
married_ever |10645223    14607038    .9479641   .2220995          0          1
(11,547,567 real changes made)
(11,547,567 real changes made)
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 2 HDFE groups                           F(   2,     30) =       3.28
Statistics robust to heteroskedasticity           Prob > F        =     0.0515
                                                  R-squared       =     0.1249
                                                  Adj R-squared   =     0.1248
                                                  Within R-sq.    =     0.0008
Number of clusters (prov)    =         31         Root MSE        =     0.2078

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
     married_ever |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0618021   .0242393    -2.55   0.016    -.1113055   -.0122988
               1  |  -.0410295   .0191583    -2.14   0.040    -.0801559   -.0019031
                  |
            _cons |   .9677682   .0079022   122.47   0.000     .9516297    .9839067
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 8 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 3 HDFE groups                           F(   2,     30) =       3.17
Statistics robust to heteroskedasticity           Prob > F        =     0.0565
                                                  R-squared       =     0.1264
                                                  Adj R-squared   =     0.1263
                                                  Within R-sq.    =     0.0008
Number of clusters (prov)    =         31         Root MSE        =     0.2076

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
     married_ever |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0657392   .0261965    -2.51   0.018    -.1192396   -.0122387
               1  |  -.0430736   .0203082    -2.12   0.042    -.0845485   -.0015987
                  |
            _cons |   .9689179   .0085118   113.83   0.000     .9515346    .9863012
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
           women#prov#year |       184         184           0    *|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 9 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 3 HDFE groups                           F(   2,     30) =       3.02
Statistics robust to heteroskedasticity           Prob > F        =     0.0640
                                                  R-squared       =     0.1257
                                                  Adj R-squared   =     0.1257
                                                  Within R-sq.    =     0.0008
Number of clusters (prov)    =         31         Root MSE        =     0.2077

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
     married_ever |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0625084   .0257823    -2.42   0.022    -.1151629    -.009854
               1  |  -.0420638   .0195038    -2.16   0.039     -.081896   -.0022317
                  |
            _cons |   .9681057    .008265   117.13   0.000     .9512262    .9849851
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#nsib |        26           0          26     |
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204     |
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 2 HDFE groups                           F(   4,     30) =       2.56
Statistics robust to heteroskedasticity           Prob > F        =     0.0590
                                                  R-squared       =     0.1251
                                                  Adj R-squared   =     0.1251
                                                  Within R-sq.    =     0.0011
Number of clusters (prov)    =         31         Root MSE        =     0.2077

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
     married_ever |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0597505   .0216235    -2.76   0.010    -.1039116   -.0155895
               1  |  -.0400699   .0179076    -2.24   0.033    -.0766421   -.0034978
                  |
women#c.sex_ratio |
               0  |  -.1443941    .056411    -2.56   0.016    -.2596008   -.0291875
               1  |  -.0708385   .0442243    -1.60   0.120    -.1611567    .0194797
                  |
            _cons |   1.078871   .0531832    20.29   0.000     .9702562    1.187486
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 7 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 3 HDFE groups                           F(   4,     30) =       2.51
Statistics robust to heteroskedasticity           Prob > F        =     0.0624
                                                  R-squared       =     0.1249
                                                  Adj R-squared   =     0.1249
                                                  Within R-sq.    =     0.0008
Number of clusters (prov)    =         31         Root MSE        =     0.2078

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
     married_ever |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0584034   .0217561    -2.68   0.012    -.1028352   -.0139716
               1  |  -.0408659   .0181891    -2.25   0.032    -.0780131   -.0037188
                  |
 women#c.lngdp_pc |
               0  |  -.0246415   .0176638    -1.40   0.173    -.0607157    .0114327
               1  |  -.0012427   .0140735    -0.09   0.930    -.0299846    .0274992
                  |
            _cons |   1.044145   .0792702    13.17   0.000      .882254    1.206036
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         4           0           4     |
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204     |
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 13 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 5 HDFE groups                           F(   6,     30) =       2.58
Statistics robust to heteroskedasticity           Prob > F        =     0.0391
                                                  R-squared       =     0.1258
                                                  Adj R-squared   =     0.1258
                                                  Within R-sq.    =     0.0009
Number of clusters (prov)    =         31         Root MSE        =     0.2077

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
     married_ever |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0644531   .0258768    -2.49   0.019    -.1173004   -.0116057
               1  |  -.0405672    .018718    -2.17   0.038    -.0787945   -.0023399
                  |
    women#c.lnsec |
               0  |  -.0046144   .0047801    -0.97   0.342    -.0143766    .0051478
               1  |   .0015676    .001608     0.97   0.337    -.0017163    .0048515
                  |
    women#c.lnpri |
               0  |   .0148983   .0116553     1.28   0.211     -.008905    .0387017
               1  |  -.0140631   .0069719    -2.02   0.053    -.0283016    .0001753
                  |
            _cons |   .9781811   .0864339    11.32   0.000     .8016595    1.154703
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
            women#lnsec_mi |         4           0           4     |
            women#lnpri_mi |         4           2           2     |
                women#nsib |        26           2          24    ?|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 2 HDFE groups                           F(   4,     30) =       4.14
Statistics robust to heteroskedasticity           Prob > F        =     0.0087
                                                  R-squared       =     0.1250
                                                  Adj R-squared   =     0.1250
                                                  Within R-sq.    =     0.0009
Number of clusters (prov)    =         31         Root MSE        =     0.2078

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
     married_ever |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0613143   .0241743    -2.54   0.017    -.1106849   -.0119438
               1  |  -.0409753   .0192622    -2.13   0.042     -.080314   -.0016365
                  |
     women#c.elig |
               0  |   .1104373   .0526225     2.10   0.044     .0029678    .2179068
               1  |   .0094826    .023766     0.40   0.693    -.0390541    .0580192
                  |
            _cons |   .9647765   .0088381   109.16   0.000     .9467267    .9828263
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 13 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 6 HDFE groups                           F(  12,     30) =       4.36
Statistics robust to heteroskedasticity           Prob > F        =     0.0005
                                                  R-squared       =     0.1263
                                                  Adj R-squared   =     0.1262
                                                  Within R-sq.    =     0.0014
Number of clusters (prov)    =         31         Root MSE        =     0.2076

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
     married_ever |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   -.058117   .0206343    -2.82   0.009    -.1002579   -.0159761
               1  |  -.0389243   .0167031    -2.33   0.027    -.0730367    -.004812
                  |
women#c.sex_ratio |
               0  |  -.1442288   .0556123    -2.59   0.015    -.2578043   -.0306532
               1  |   -.074401   .0427624    -1.74   0.092    -.1617335    .0129315
                  |
 women#c.lngdp_pc |
               0  |  -.0213438   .0161685    -1.32   0.197    -.0543643    .0116767
               1  |  -.0024795   .0112467    -0.22   0.827    -.0254483    .0204892
                  |
    women#c.lnsec |
               0  |  -.0013179    .003493    -0.38   0.709    -.0084515    .0058158
               1  |   .0022879   .0015561     1.47   0.152    -.0008901    .0054658
                  |
    women#c.lnpri |
               0  |   .0081008   .0079804     1.02   0.318    -.0081973    .0243988
               1  |  -.0155135   .0069755    -2.22   0.034    -.0297594   -.0012676
                  |
     women#c.elig |
               0  |   .1142216   .0464032     2.46   0.020     .0194536    .2089896
               1  |   .0246909    .024257     1.02   0.317    -.0248485    .0742303
                  |
            _cons |   1.184524    .159213     7.44   0.000     .8593681    1.509681
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         4           0           4     |
            women#lnsec_mi |         4           2           2     |
            women#lnpri_mi |         4           4           0    ?|
                women#nsib |        26           2          24    ?|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 15 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 7 HDFE groups                           F(  12,     30) =       4.34
Statistics robust to heteroskedasticity           Prob > F        =     0.0005
                                                  R-squared       =     0.1278
                                                  Adj R-squared   =     0.1277
                                                  Within R-sq.    =     0.0015
Number of clusters (prov)    =         31         Root MSE        =     0.2074

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
     married_ever |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0615484   .0221051    -2.78   0.009     -.106693   -.0164038
               1  |  -.0405763   .0175762    -2.31   0.028    -.0764717   -.0046808
                  |
women#c.sex_ratio |
               0  |  -.1340013   .0524216    -2.56   0.016    -.2410605   -.0269421
               1  |  -.0731389   .0426142    -1.72   0.096    -.1601686    .0138909
                  |
 women#c.lngdp_pc |
               0  |  -.0237961   .0169223    -1.41   0.170    -.0583561    .0107639
               1  |  -.0046298   .0124727    -0.37   0.713    -.0301025    .0208429
                  |
    women#c.lnsec |
               0  |  -.0008176   .0036866    -0.22   0.826    -.0083467    .0067115
               1  |   .0028921   .0015755     1.84   0.076    -.0003256    .0061097
                  |
    women#c.lnpri |
               0  |   .0052056   .0085677     0.61   0.548    -.0122919    .0227032
               1  |  -.0165219   .0074453    -2.22   0.034    -.0317273   -.0013166
                  |
     women#c.elig |
               0  |   .1228301   .0485109     2.53   0.017     .0237575    .2219026
               1  |   .0272338   .0256863     1.06   0.297    -.0252247    .0796923
                  |
            _cons |   1.209937    .177751     6.81   0.000     .8469212    1.572953
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         4           0           4     |
            women#lnsec_mi |         4           2           2     |
            women#lnpri_mi |         4           4           0    ?|
                women#nsib |        26           2          24    ?|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204    ?|
           women#prov#year |       184         184           0    *|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
   late_marr |10645223    14607038    .8071595   .3945289          0          1
(11,547,567 real changes made)
(11,547,567 real changes made)
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 2 HDFE groups                           F(   2,     30) =       1.96
Statistics robust to heteroskedasticity           Prob > F        =     0.1586
                                                  R-squared       =     0.1413
                                                  Adj R-squared   =     0.1413
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.3656

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
        late_marr |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0312038   .0174973    -1.78   0.085    -.0669381    .0045306
               1  |  -.0292088   .0155016    -1.88   0.069    -.0608673    .0024497
                  |
            _cons |   .8188276   .0059167   138.39   0.000     .8067441    .8309111
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 8 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 3 HDFE groups                           F(   2,     30) =       1.66
Statistics robust to heteroskedasticity           Prob > F        =     0.2077
                                                  R-squared       =     0.1448
                                                  Adj R-squared   =     0.1448
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.3648

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
        late_marr |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0321165    .018745    -1.71   0.097    -.0703989    .0061659
               1  |  -.0279166   .0165634    -1.69   0.102    -.0617437    .0059105
                  |
            _cons |   .8187471   .0063652   128.63   0.000     .8057476    .8317466
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
           women#prov#year |       184         184           0    *|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 9 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 3 HDFE groups                           F(   2,     30) =       2.56
Statistics robust to heteroskedasticity           Prob > F        =     0.0945
                                                  R-squared       =     0.1420
                                                  Adj R-squared   =     0.1419
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.3655

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
        late_marr |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0325201   .0174763    -1.86   0.073    -.0682115    .0031713
               1  |  -.0311689   .0141315    -2.21   0.035    -.0600293   -.0023085
                  |
            _cons |   .8194629   .0055901   146.59   0.000     .8080463    .8308795
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#nsib |        26           0          26     |
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204     |
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 2 HDFE groups                           F(   4,     30) =       2.97
Statistics robust to heteroskedasticity           Prob > F        =     0.0353
                                                  R-squared       =     0.1414
                                                  Adj R-squared   =     0.1413
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.3656

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
        late_marr |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0295712   .0166082    -1.78   0.085    -.0634898    .0043473
               1  |  -.0289708   .0157495    -1.84   0.076    -.0611356     .003194
                  |
women#c.sex_ratio |
               0  |  -.1148978   .0397551    -2.89   0.007    -.1960885   -.0337071
               1  |  -.0175707   .0494625    -0.36   0.725    -.1185866    .0834452
                  |
            _cons |   .8876004   .0365124    24.31   0.000      .813032    .9621687
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 7 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 3 HDFE groups                           F(   4,     30) =       1.94
Statistics robust to heteroskedasticity           Prob > F        =     0.1292
                                                  R-squared       =     0.1414
                                                  Adj R-squared   =     0.1413
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.3656

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
        late_marr |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0333424   .0166406    -2.00   0.054    -.0673269    .0006422
               1  |  -.0322016   .0152907    -2.11   0.044    -.0634294   -.0009738
                  |
 women#c.lngdp_pc |
               0  |   .0140739   .0282975     0.50   0.623    -.0437172     .071865
               1  |    .021547   .0161718     1.33   0.193    -.0114802    .0545742
                  |
            _cons |   .7148718   .1242124     5.76   0.000     .4611962    .9685474
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         4           0           4     |
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204     |
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 13 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 5 HDFE groups                           F(   6,     30) =       1.64
Statistics robust to heteroskedasticity           Prob > F        =     0.1695
                                                  R-squared       =     0.1420
                                                  Adj R-squared   =     0.1420
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.3654

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
        late_marr |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0332019   .0178699    -1.86   0.073    -.0696971    .0032933
               1  |  -.0277292    .013196    -2.10   0.044     -.054679   -.0007795
                  |
    women#c.lnsec |
               0  |   .0016644   .0084341     0.20   0.845    -.0155603     .018889
               1  |   .0014771   .0045108     0.33   0.746    -.0077351    .0106893
                  |
    women#c.lnpri |
               0  |   .0049782   .0141189     0.35   0.727    -.0238564    .0338129
               1  |   -.033135   .0151764    -2.18   0.037    -.0641294   -.0021407
                  |
            _cons |   .9649526   .1142519     8.45   0.000      .731619    1.198286
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
            women#lnsec_mi |         4           0           4     |
            women#lnpri_mi |         4           2           2     |
                women#nsib |        26           2          24    ?|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 2 HDFE groups                           F(   4,     30) =       2.50
Statistics robust to heteroskedasticity           Prob > F        =     0.0636
                                                  R-squared       =     0.1414
                                                  Adj R-squared   =     0.1413
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.3656

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
        late_marr |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0308145   .0173892    -1.77   0.087     -.066328     .004699
               1  |  -.0291664   .0155534    -1.88   0.071    -.0609307    .0025979
                  |
     women#c.elig |
               0  |    .088126   .0567471     1.55   0.131    -.0277671    .2040191
               1  |   .0074105   .0258807     0.29   0.777     -.045445    .0602659
                  |
            _cons |   .8164445   .0061858   131.99   0.000     .8038114    .8290776
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 13 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 6 HDFE groups                           F(  12,     30) =       2.95
Statistics robust to heteroskedasticity           Prob > F        =     0.0079
                                                  R-squared       =     0.1421
                                                  Adj R-squared   =     0.1421
                                                  Within R-sq.    =     0.0003
Number of clusters (prov)    =         31         Root MSE        =     0.3654

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
        late_marr |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   -.032267   .0148376    -2.17   0.038    -.0625694   -.0019646
               1  |  -.0291211    .012885    -2.26   0.031    -.0554358   -.0028063
                  |
women#c.sex_ratio |
               0  |  -.1276867    .043908    -2.91   0.007    -.2173588   -.0380146
               1  |  -.0290764   .0490535    -0.59   0.558     -.129257    .0711042
                  |
 women#c.lngdp_pc |
               0  |   .0134924   .0274616     0.49   0.627    -.0425917    .0695766
               1  |   .0151542   .0174055     0.87   0.391    -.0203926    .0507011
                  |
    women#c.lnsec |
               0  |   .0041439   .0072584     0.57   0.572    -.0106798    .0189676
               1  |   .0020517   .0041824     0.49   0.627    -.0064899    .0105932
                  |
    women#c.lnpri |
               0  |   .0004697   .0128796     0.04   0.971    -.0258339    .0267733
               1  |  -.0339251   .0153967    -2.20   0.035    -.0653693   -.0024809
                  |
     women#c.elig |
               0  |   .1084105   .0463092     2.34   0.026     .0138346    .2029864
               1  |   .0341902   .0206166     1.66   0.108    -.0079145     .076295
                  |
            _cons |   .9735137   .1803718     5.40   0.000     .6051454    1.341882
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         4           0           4     |
            women#lnsec_mi |         4           2           2     |
            women#lnpri_mi |         4           4           0    ?|
                women#nsib |        26           2          24    ?|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 15 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 7 HDFE groups                           F(  12,     30) =       3.45
Statistics robust to heteroskedasticity           Prob > F        =     0.0029
                                                  R-squared       =     0.1457
                                                  Adj R-squared   =     0.1456
                                                  Within R-sq.    =     0.0003
Number of clusters (prov)    =         31         Root MSE        =     0.3647

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
        late_marr |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0331698   .0155832    -2.13   0.042    -.0649949   -.0013447
               1  |  -.0273693   .0136875    -2.00   0.055    -.0553229    .0005844
                  |
women#c.sex_ratio |
               0  |  -.1574778   .0469356    -3.36   0.002    -.2533332   -.0616225
               1  |  -.0410627   .0478925    -0.86   0.398    -.1388722    .0567469
                  |
 women#c.lngdp_pc |
               0  |   .0138324   .0306956     0.45   0.655    -.0488564    .0765212
               1  |   .0093251   .0205556     0.45   0.653    -.0326551    .0513053
                  |
    women#c.lnsec |
               0  |   .0053167   .0069039     0.77   0.447    -.0087829    .0194164
               1  |  -.0002814   .0043666    -0.06   0.949    -.0091993    .0086364
                  |
    women#c.lnpri |
               0  |   .0032731   .0121075     0.27   0.789    -.0214536    .0279998
               1  |   -.031134   .0150916    -2.06   0.048    -.0619552   -.0003129
                  |
     women#c.elig |
               0  |   .0979425   .0456482     2.15   0.040     .0047164    .1911686
               1  |   .0272771    .020586     1.33   0.195    -.0147653    .0693194
                  |
            _cons |   .9846836   .1919871     5.13   0.000     .5925936    1.376774
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         4           0           4     |
            women#lnsec_mi |         4           2           2     |
            women#lnpri_mi |         4           4           0    ?|
                women#nsib |        26           2          24    ?|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204    ?|
           women#prov#year |       184         184           0    *|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
        work |10645223    14607038     .835665   .3705793          0          1
(11,547,567 real changes made)
(11,547,567 real changes made)
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 2 HDFE groups                           F(   2,     30) =       1.49
Statistics robust to heteroskedasticity           Prob > F        =     0.2412
                                                  R-squared       =     0.1757
                                                  Adj R-squared   =     0.1757
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =     0.3365

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
             work |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0116817    .009497    -1.23   0.228    -.0310772    .0077138
               1  |  -.0031001   .0124815    -0.25   0.806    -.0285906    .0223904
                  |
            _cons |   .8384935   .0040408   207.51   0.000     .8302412    .8467459
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 8 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 3 HDFE groups                           F(   2,     30) =       0.26
Statistics robust to heteroskedasticity           Prob > F        =     0.7694
                                                  R-squared       =     0.1827
                                                  Adj R-squared   =     0.1827
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =     0.3350

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
             work |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0020228   .0071886    -0.28   0.780    -.0167039    .0126583
               1  |   .0031466   .0108586     0.29   0.774    -.0190296    .0253227
                  |
            _cons |    .835431   .0032326   258.44   0.000     .8288291    .8420328
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
           women#prov#year |       184         184           0    *|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 9 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 3 HDFE groups                           F(   2,     30) =       1.76
Statistics robust to heteroskedasticity           Prob > F        =     0.1896
                                                  R-squared       =     0.1758
                                                  Adj R-squared   =     0.1757
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =     0.3364

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
             work |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0120365   .0086021    -1.40   0.172    -.0296044    .0055313
               1  |   -.002873   .0125318    -0.23   0.820    -.0284664    .0227204
                  |
            _cons |   .8385163   .0038457   218.04   0.000     .8306622    .8463703
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#nsib |        26           0          26     |
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204     |
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 2 HDFE groups                           F(   4,     30) =       1.78
Statistics robust to heteroskedasticity           Prob > F        =     0.1584
                                                  R-squared       =     0.1757
                                                  Adj R-squared   =     0.1757
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =     0.3364

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
             work |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0119444   .0095358    -1.25   0.220     -.031419    .0075302
               1  |  -.0041248   .0120813    -0.34   0.735    -.0287982    .0205485
                  |
women#c.sex_ratio |
               0  |   .0184914   .0134665     1.37   0.180    -.0090109    .0459936
               1  |   .0756471   .0333368     2.27   0.031     .0075642      .14373
                  |
            _cons |   .7905734   .0195689    40.40   0.000     .7506084    .8305383
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 8 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 3 HDFE groups                           F(   4,     30) =       1.04
Statistics robust to heteroskedasticity           Prob > F        =     0.4035
                                                  R-squared       =     0.1757
                                                  Adj R-squared   =     0.1757
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =     0.3365

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
             work |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0130584   .0092937    -1.41   0.170    -.0320388    .0059219
               1  |  -.0050304   .0119935    -0.42   0.678    -.0295244    .0194635
                  |
 women#c.lngdp_pc |
               0  |   .0092914   .0092341     1.01   0.322     -.009567    .0281499
               1  |   .0137791    .014453     0.95   0.348    -.0157379    .0432961
                  |
            _cons |   .7711518   .0608608    12.67   0.000     .6468574    .8954461
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         4           0           4     |
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204     |
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 13 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 5 HDFE groups                           F(   6,     30) =       1.88
Statistics robust to heteroskedasticity           Prob > F        =     0.1164
                                                  R-squared       =     0.1758
                                                  Adj R-squared   =     0.1758
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =     0.3364

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
             work |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0134264   .0091553    -1.47   0.153    -.0321239    .0052712
               1  |  -.0013908   .0128046    -0.11   0.914    -.0275412    .0247597
                  |
    women#c.lnsec |
               0  |   .0006898   .0031854     0.22   0.830    -.0058156    .0071953
               1  |  -.0003798   .0049772    -0.08   0.940    -.0105446     .009785
                  |
    women#c.lnpri |
               0  |   .0135885   .0047426     2.87   0.008     .0039028    .0232742
               1  |  -.0132862    .010263    -1.29   0.205    -.0342461    .0076737
                  |
            _cons |   .8331992   .0597097    13.95   0.000     .7112556    .9551427
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
            women#lnsec_mi |         4           0           4     |
            women#lnpri_mi |         4           2           2     |
                women#nsib |        26           2          24    ?|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 2 HDFE groups                           F(   4,     30) =       0.80
Statistics robust to heteroskedasticity           Prob > F        =     0.5340
                                                  R-squared       =     0.1757
                                                  Adj R-squared   =     0.1757
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =     0.3365

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
             work |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |  -.0117097   .0094391    -1.24   0.224    -.0309868    .0075675
               1  |  -.0031101   .0124439    -0.25   0.804    -.0285239    .0223036
                  |
     women#c.elig |
               0  |  -.0063366   .0206707    -0.31   0.761    -.0485518    .0358787
               1  |  -.0017545   .0352895    -0.05   0.961    -.0738252    .0703161
                  |
            _cons |   .8386979   .0040007   209.64   0.000     .8305274    .8468684
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 13 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 6 HDFE groups                           F(  12,     30) =       1.22
Statistics robust to heteroskedasticity           Prob > F        =     0.3155
                                                  R-squared       =     0.1758
                                                  Adj R-squared   =     0.1758
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.3364

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
             work |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   -.014801   .0091777    -1.61   0.117    -.0335444    .0039424
               1  |  -.0039507   .0119375    -0.33   0.743    -.0283304    .0204289
                  |
women#c.sex_ratio |
               0  |   .0148774   .0144959     1.03   0.313    -.0147273     .044482
               1  |   .0721273   .0328341     2.20   0.036     .0050712    .1391835
                  |
 women#c.lngdp_pc |
               0  |   .0075368   .0091455     0.82   0.416    -.0111408    .0262145
               1  |   .0110032   .0138221     0.80   0.432    -.0172253    .0392318
                  |
    women#c.lnsec |
               0  |   .0004028   .0030128     0.13   0.895    -.0057501    .0065557
               1  |  -.0006536   .0050527    -0.13   0.898    -.0109727    .0096655
                  |
    women#c.lnpri |
               0  |   .0143266   .0048206     2.97   0.006     .0044816    .0241715
               1  |  -.0124838   .0097211    -1.28   0.209     -.032337    .0073695
                  |
     women#c.elig |
               0  |  -.0053995   .0211514    -0.26   0.800    -.0485963    .0377974
               1  |   .0000585   .0370969     0.00   0.999    -.0757036    .0758205
                  |
            _cons |    .728946   .0849689     8.58   0.000     .5554163    .9024758
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         4           0           4     |
            women#lnsec_mi |         4           2           2     |
            women#lnpri_mi |         4           4           0    ?|
                women#nsib |        26           2          24    ?|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 15 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 7 HDFE groups                           F(  12,     30) =       1.59
Statistics robust to heteroskedasticity           Prob > F        =     0.1463
                                                  R-squared       =     0.1828
                                                  Adj R-squared   =     0.1827
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =     0.3350

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
             work |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   -.004029   .0066432    -0.61   0.549    -.0175963    .0095383
               1  |   .0047399   .0104301     0.45   0.653    -.0165612     .026041
                  |
women#c.sex_ratio |
               0  |  -.0004009    .010818    -0.04   0.971    -.0224942    .0216925
               1  |   .0418577   .0275328     1.52   0.139    -.0143717    .0980871
                  |
 women#c.lngdp_pc |
               0  |   .0008489   .0067751     0.13   0.901    -.0129876    .0146855
               1  |  -.0041093   .0136478    -0.30   0.765    -.0319819    .0237633
                  |
    women#c.lnsec |
               0  |   .0010344   .0017585     0.59   0.561     -.002557    .0046258
               1  |  -.0001905   .0039238    -0.05   0.962    -.0082041     .007823
                  |
    women#c.lnpri |
               0  |   .0146195   .0048621     3.01   0.005     .0046898    .0245493
               1  |  -.0121305   .0079003    -1.54   0.135     -.028265    .0040041
                  |
     women#c.elig |
               0  |   .0034015   .0144196     0.24   0.815    -.0260473    .0328503
               1  |   .0021732   .0311949     0.07   0.945    -.0615352    .0658816
                  |
            _cons |   .8027342    .079371    10.11   0.000      .640637    .9648313
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         4           0           4     |
            women#lnsec_mi |         4           2           2     |
            women#lnpri_mi |         4           4           0    ?|
                women#nsib |        26           2          24    ?|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204    ?|
           women#prov#year |       184         184           0    *|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)
(11,547,567 missing values generated)

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
    high_occ |10645223    14607038    .1074155   .3096409          0          1
(11,547,567 real changes made)
(11,547,567 real changes made)
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 2 HDFE groups                           F(   2,     30) =       4.04
Statistics robust to heteroskedasticity           Prob > F        =     0.0279
                                                  R-squared       =     0.0275
                                                  Adj R-squared   =     0.0274
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =     0.3054

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
         high_occ |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0157543   .0055749     2.83   0.008     .0043687    .0271398
               1  |   .0166601   .0093083     1.79   0.084      -.00235    .0356702
                  |
            _cons |   .1011485   .0025603    39.51   0.000     .0959198    .1063773
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 8 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 3 HDFE groups                           F(   2,     30) =       8.15
Statistics robust to heteroskedasticity           Prob > F        =     0.0015
                                                  R-squared       =     0.0291
                                                  Adj R-squared   =     0.0291
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =     0.3051

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
         high_occ |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0172403   .0045903     3.76   0.001     .0078656     .026615
               1  |   .0191463   .0074243     2.58   0.015     .0039839    .0343087
                  |
            _cons |   .1003776    .001919    52.31   0.000     .0964586    .1042966
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
           women#prov#year |       184         184           0    *|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 9 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 3 HDFE groups                           F(   2,     30) =       2.77
Statistics robust to heteroskedasticity           Prob > F        =     0.0785
                                                  R-squared       =     0.0288
                                                  Adj R-squared   =     0.0287
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =     0.3052

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
         high_occ |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0172843   .0076873     2.25   0.032     .0015848    .0329838
               1  |   .0192623   .0131589     1.46   0.154    -.0076118    .0461364
                  |
            _cons |   .1003465    .003881    25.86   0.000     .0924204    .1082726
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#nsib |        26           0          26     |
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204     |
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 2 HDFE groups                           F(   4,     30) =       2.48
Statistics robust to heteroskedasticity           Prob > F        =     0.0648
                                                  R-squared       =     0.0275
                                                  Adj R-squared   =     0.0274
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =     0.3054

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
         high_occ |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0157288   .0056374     2.79   0.009     .0042156     .027242
               1  |   .0164746   .0092514     1.78   0.085    -.0024192    .0353685
                  |
women#c.sex_ratio |
               0  |   .0017912   .0174523     0.10   0.919    -.0338512    .0374336
               1  |   .0136896   .0155914     0.88   0.387    -.0181523    .0455315
                  |
            _cons |   .0932871   .0146173     6.38   0.000     .0634346    .1231395
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 7 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 3 HDFE groups                           F(   4,     30) =       3.34
Statistics robust to heteroskedasticity           Prob > F        =     0.0224
                                                  R-squared       =     0.0275
                                                  Adj R-squared   =     0.0274
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =     0.3054

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
         high_occ |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0168177   .0052971     3.17   0.003     .0059996    .0276359
               1  |   .0163087   .0090614     1.80   0.082    -.0021972    .0348146
                  |
 women#c.lngdp_pc |
               0  |  -.0078016   .0132325    -0.59   0.560    -.0348259    .0192227
               1  |   .0023376   .0096278     0.24   0.810     -.017325    .0220001
                  |
            _cons |   .1174244   .0646256     1.82   0.079    -.0145588    .2494075
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         4           0           4     |
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204     |
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 13 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 5 HDFE groups                           F(   6,     30) =       2.30
Statistics robust to heteroskedasticity           Prob > F        =     0.0605
                                                  R-squared       =     0.0288
                                                  Adj R-squared   =     0.0288
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.3052

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
         high_occ |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0155368   .0073215     2.12   0.042     .0005844    .0304892
               1  |   .0188605   .0130562     1.44   0.159    -.0078038    .0455247
                  |
    women#c.lnsec |
               0  |   -.002426    .002758    -0.88   0.386    -.0080587    .0032066
               1  |  -.0010852   .0017194    -0.63   0.533    -.0045966    .0024262
                  |
    women#c.lnpri |
               0  |   .0152008   .0070366     2.16   0.039     .0008301    .0295715
               1  |   .0027653   .0046938     0.59   0.560    -.0068207    .0123512
                  |
            _cons |   .0133615     .05365     0.25   0.805    -.0962063    .1229293
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
            women#lnsec_mi |         4           0           4     |
            women#lnpri_mi |         4           2           2     |
                women#nsib |        26           2          24    ?|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   = 10,645,223
Absorbing 2 HDFE groups                           F(   4,     30) =       2.78
Statistics robust to heteroskedasticity           Prob > F        =     0.0448
                                                  R-squared       =     0.0275
                                                  Adj R-squared   =     0.0274
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =     0.3054

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
         high_occ |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0157709   .0055874     2.82   0.008     .0043598     .027182
               1  |    .016646   .0093122     1.79   0.084    -.0023721     .035664
                  |
     women#c.elig |
               0  |   .0037657   .0164159     0.23   0.820      -.02976    .0372915
               1  |  -.0024669    .018265    -0.14   0.893    -.0397689    .0348352
                  |
            _cons |   .1011219   .0028121    35.96   0.000     .0953789    .1068649
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62     |
     women#year#year_birth |       206           0         206    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 13 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 6 HDFE groups                           F(  12,     30) =       1.85
Statistics robust to heteroskedasticity           Prob > F        =     0.0848
                                                  R-squared       =     0.0288
                                                  Adj R-squared   =     0.0288
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.3052

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
         high_occ |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0153196   .0065453     2.34   0.026     .0019523    .0286868
               1  |   .0170882   .0120806     1.41   0.168    -.0075836      .04176
                  |
women#c.sex_ratio |
               0  |   .0087083   .0146626     0.59   0.557    -.0212368    .0386533
               1  |   .0180868   .0155623     1.16   0.254    -.0136955    .0498692
                  |
 women#c.lngdp_pc |
               0  |  -.0004825   .0122571    -0.04   0.969    -.0255148    .0245497
               1  |   .0091867   .0100579     0.91   0.368    -.0113543    .0297278
                  |
    women#c.lnsec |
               0  |  -.0028321   .0027959    -1.01   0.319    -.0085421     .002878
               1  |  -.0017663   .0015645    -1.13   0.268    -.0049615     .001429
                  |
    women#c.lnpri |
               0  |   .0159191   .0072935     2.18   0.037     .0010238    .0308143
               1  |   .0041949   .0045743     0.92   0.366    -.0051471    .0135369
                  |
     women#c.elig |
               0  |  -.0168465   .0148885    -1.13   0.267    -.0472528    .0135598
               1  |   -.019857   .0182159    -1.09   0.284    -.0570589    .0173448
                  |
            _cons |  -.0312478   .0951756    -0.33   0.745    -.2256223    .1631266
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         4           0           4     |
            women#lnsec_mi |         4           2           2     |
            women#lnpri_mi |         4           4           0    ?|
                women#nsib |        26           2          24    ?|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 15 iterations)

HDFE Linear regression                            Number of obs   = 10,645,222
Absorbing 7 HDFE groups                           F(  12,     30) =       3.27
Statistics robust to heteroskedasticity           Prob > F        =     0.0042
                                                  R-squared       =     0.0305
                                                  Adj R-squared   =     0.0304
                                                  Within R-sq.    =     0.0001
Number of clusters (prov)    =         31         Root MSE        =     0.3049

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
         high_occ |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               0  |   .0170117   .0044165     3.85   0.001     .0079921    .0260313
               1  |   .0200831     .01002     2.00   0.054    -.0003804    .0405466
                  |
women#c.sex_ratio |
               0  |   .0121788   .0121539     1.00   0.324    -.0126428    .0370005
               1  |   .0146657    .015194     0.97   0.342    -.0163645    .0456959
                  |
 women#c.lngdp_pc |
               0  |  -.0007792   .0105809    -0.07   0.942    -.0223882    .0208299
               1  |   .0063063   .0089383     0.71   0.486    -.0119482    .0245607
                  |
    women#c.lnsec |
               0  |  -.0020168   .0024388    -0.83   0.415    -.0069974    .0029638
               1  |  -.0009819   .0014476    -0.68   0.503    -.0039383    .0019746
                  |
    women#c.lnpri |
               0  |   .0147344   .0066843     2.20   0.035     .0010833    .0283855
               1  |   .0042305   .0045404     0.93   0.359    -.0050423    .0135033
                  |
     women#c.elig |
               0  |  -.0183513   .0124332    -1.48   0.150    -.0437432    .0070406
               1  |  -.0185948   .0152122    -1.22   0.231    -.0496622    .0124726
                  |
            _cons |  -.0247335    .086989    -0.28   0.778    -.2023887    .1529218
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         4           0           4     |
            women#lnsec_mi |         4           2           2     |
            women#lnpri_mi |         4           4           0    ?|
                women#nsib |        26           2          24    ?|
                women#prov |        62          62           0    *|
   women#prov#c.year_birth |        62           0          62    ?|
     women#year#year_birth |       206           2         204    ?|
           women#prov#year |       184         184           0    *|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_census_control.xls
dir : seeout

. 
. keep coef_* se_* mean_* std_*  

. drop if mi(coef_men_senior)
(11,547,558 observations deleted)

. gen pos = _n

. 
. save "$path3/coef_census_ind_control",replace 
(note: file /Users/Wei/Dropbox/Fertility/Results/coef_census_ind_control.dta not found)
file /Users/Wei/Dropbox/Fertility/Results/coef_census_ind_control.dta saved

.  
. 
. set more off

. 
. use "$path2/marr_policy", clear

. gen mort_rate = 100 -sur_rate if women == 1 
(6,645,526 missing values generated)

. ren year yr_t

. gen year = year_birth +15

. merge m:1 prov year using "$path2/fines_economic_new", keepusing(lngdp_pc secondary primary)
(note: variable prov was byte, now float to accommodate using data's values)

    Result                           # of obs.
    -----------------------------------------
    not matched                           589
        from master                         0  (_merge==1)
        from using                        589  (_merge==2)

    matched                        11,547,566  (_merge==3)
    -----------------------------------------

. drop if _merge == 2 
(589 observations deleted)

. drop _merge 

. drop year

. ren yr_t year

. egen male_prop = mean(men), by(prov year_birth)

. gen sex_ratio = male_prop/(1-male_prop)

. 
. 
. egen fine_6_20 = rowmean(fine_age6-fine_age20)

. drop fine_age*

. gl FINE_ALL = "fine_6_15 fine_16_20 fine_21_25 fine_26_30"

. gl FINE_VAR_4 =  "fine_6_15 fine_16_20 fine_21_30"

. 
. gl FINE_VAR_3 =  "fine_6_15 fine_16_20 fine_21_25"

. gl FINE_VAR_1 = "fine_6_20"

. 
. replace nbroth = 99 if mi(nbroth)
(11,325,299 real changes made)

. replace nsister = 99 if mi(nsister)
(11,325,299 real changes made)

. 
. gen lnsec = ln(secondary)
(577,564 missing values generated)

. gen lnpri = ln(primary)
(352,981 missing values generated)

. gen lnsec_mi = mi(lnsec)

. gen lnpri_mi = mi(lnpri)

. gen lngdp_pc_mi = mi(lnpri)

. 
. egen mean_lnsec = mean(lnsec)

. egen mean_lnpri = mean(lnpri)

. egen mean_lngdp = mean(lngdp_pc)

. 
. replace lnsec = mean_lnsec if mi(lnsec)
(577,564 real changes made)

. replace lnpri = mean_lnpri if mi(lnpri)
(352,981 real changes made)

. replace lngdp_pc = mean_lngdp if mi(lngdp_pc)
(41,708 real changes made)

. drop mean_lnsec  mean_lnpri mean_lngdp

. 
. gen provid = prov 

. merge m:1 provid using "$path2/Comp_Law", keepusing(law_*) nogen 

    Result                           # of obs.
    -----------------------------------------
    not matched                             1
        from master                         0  
        from using                          1  

    matched                        11,547,566  
    -----------------------------------------

. 
. cap drop elig

. gen elig = 1 if year_birth >= law_year - 6 
(11,539,033 missing values generated)

. replace elig = 0 if year_birth <= law_year - 16
(10,501,045 real changes made)

. replace elig = (16-(law_year - year_birth ))/10 if law_year - 16<year_birth & year_birth<law_ye
> ar - 6 
(1,037,988 real changes made)

. drop provid

. 
. 
. gl CONTROL_1 = "women#prov##c.year_birth women#year#year_birth"

. gl CONTROL_2 = "women#prov##c.year_birth women#year#year_birth women#prov#year"

. 
. gen urban_hk = hktype == 2 

. gen urban_res = city_flag <= 2 if !mi(city_flag)
(4,140,909 missing values generated)

. replace urban_res = urban_hk if  mi(urban_res)
(4,140,909 real changes made)

. set more off 

. replace high_occ = 0 if high_occ == . & !mi(work)
(1,603,442 real changes made)

. 
. egen han_p = mean(han), by(year hhid prov region)
(1 missing value generated)

. 
. 
. drop if han_p >0 & han_p < 1 // drop mixed households
(325,664 observations deleted)

. drop if women == 0 
(5,705,234 observations deleted)

. set more off

.  
. set more off

. gen nsib = nbroth + nsister
(1 missing value generated)

. gen n_death = n_birth - n_survive
(567,308 missing values generated)

. 
. replace n_death = . if n_death < 0
(1 real change made, 1 to missing)

. replace n_birth = 0 if married_ever == 0 
(70,329 real changes made)

. replace n_mbirth = 0 if married_ever == 0  
(70,244 real changes made)

. replace n_fbirth = 0 if married_ever == 0  
(70,225 real changes made)

. replace n_survive = 0 if married_ever == 0 
(70,299 real changes made)

. replace n_mchild = 0 if married_ever == 0 
(70,225 real changes made)

. replace n_fchild = 0 if married_ever == 0 
(70,204 real changes made)

. drop if mi(n_birth)
(494,418 observations deleted)

. drop if mi(n_death)
(72,891 observations deleted)

. 
. 
. cap erase "$path3/OCP_women_birth_control.xls"

. cap erase "$path3/OCP_women_birth_control.txt"

. foreach y_var in "n_birth" {
  2. gen coef_women_`y_var' = .
  3. gen se_women_`y_var' = .
  4. gen mean_`y_var'  = . 
  5. gen std_`y_var'  = . 
  6. su `y_var'  [aw = wt] if han 
  7. replace mean_`y_var' = `r(mean)'  
  8. replace std_`y_var' = `r(sd)'  
  9. local j = 1 
 10. 
. foreach x_var in  "$FINE_VAR_1"{
 11. reghdfe `y_var' women#c.(`x_var') if han [aw = wt], a($CONTROL_1) cluster(prov)
 12. replace coef_women_`y_var' = _b[1.women#`x_var']   in `j'
 13. replace se_women_`y_var' = _se[1.women#`x_var']    in `j'
 14. outreg2 using "$path3/OCP_women_birth_control.xls", dec(3) append
 15. local ++j
 16. 
. reghdfe `y_var' women#c.(`x_var') if han  [aw = wt], a($CONTROL_1 women#prov#year) cluster(prov
> )
 17. replace coef_women_`y_var' = _b[1.women#`x_var']    in `j'
 18. replace se_women_`y_var' = _se[1.women#`x_var']    in `j'
 19. outreg2 using "$path3/OCP_women_birth_control.xls", dec(3) append
 20. local ++j
 21. 
. reghdfe `y_var' women#c.(`x_var') if han  [aw = wt], a(women#nsib $CONTROL_1) cluster(prov)
 22. replace coef_women_`y_var' = _b[1.women#`x_var']   in `j'
 23. replace se_women_`y_var' = _se[1.women#`x_var']   in `j'
 24. outreg2 using "$path3/OCP_women_birth_control.xls", dec(3) append
 25. local ++j
 26. 
. reghdfe `y_var' women#c.(`x_var' sex_ratio) if han  [aw = wt], a($CONTROL_1) cluster(prov)
 27. replace coef_women_`y_var' = _b[1.women#`x_var']   in `j'
 28. replace se_women_`y_var' = _se[1.women#`x_var']   in `j'
 29. outreg2 using "$path3/OCP_women_birth_control.xls", dec(3) append
 30. local ++j
 31. 
. reghdfe `y_var' women#c.(`x_var' lngdp_pc) if han  [aw = wt], a(women#lngdp_pc_mi $CONTROL_1) c
> luster(prov)
 32. replace coef_women_`y_var' = _b[1.women#`x_var']    in `j'
 33. replace se_women_`y_var' = _se[1.women#`x_var']  in `j'
 34. outreg2 using "$path3/OCP_women_birth_control.xls", dec(3) append
 35. local ++j
 36. 
. reghdfe `y_var' women#c.(`x_var' lnsec lnpri) if han  [aw = wt], a(women#lnsec_mi  women#lnpri_
> mi  $CONTROL_1) cluster(prov)
 37. replace coef_women_`y_var' = _b[1.women#`x_var']   in `j'
 38. replace se_women_`y_var' = _se[1.women#`x_var']    in `j'
 39. outreg2 using "$path3/OCP_women_birth_control.xls", dec(3) append
 40. local ++j
 41. 
. reghdfe `y_var' women#c.(`x_var' elig) if han  [aw = wt], a( $CONTROL_1) cluster(prov)
 42. replace coef_women_`y_var' = _b[1.women#`x_var']    in `j'
 43. replace se_women_`y_var' = _se[1.women#`x_var']   in `j'
 44. outreg2 using "$path3/OCP_women_birth_control.xls", dec(3) append
 45. local ++j
 46. 
. 
. reghdfe `y_var' women#c.(`x_var' lngdp_pc lnsec lnpri elig sex_ratio) if han [aw = wt], a(women
> #lnsec_mi  women#lnpri_mi women#lngdp_pc_mi women#nsib $CONTROL_1 women#prov#year) cluster(prov
> )
 47. replace coef_women_`y_var' = _b[1.women#`x_var']   in `j'
 48. replace se_women_`y_var' = _se[1.women#`x_var']  in `j'
 49. outreg2 using "$path3/OCP_women_birth_control.xls", dec(3) append
 50. local ++j
 51. }
 52. }
(4,949,360 missing values generated)
(4,949,360 missing values generated)
(4,949,360 missing values generated)
(4,949,360 missing values generated)

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
     n_birth | 4621938     6514068     1.99264   1.202118          0         15
(4,949,360 real changes made)
(4,949,360 real changes made)
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   =  4,621,938
Absorbing 2 HDFE groups                           F(   1,     30) =       3.05
Statistics robust to heteroskedasticity           Prob > F        =     0.0911
                                                  R-squared       =     0.4181
                                                  Adj R-squared   =     0.4181
                                                  Within R-sq.    =     0.0005
Number of clusters (prov)    =         31         Root MSE        =     0.9170

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
          n_birth |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               1  |  -.2021068   .1157752    -1.75   0.091    -.4385512    .0343376
                  |
            _cons |   2.074815   .0470732    44.08   0.000     1.978679    2.170951
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        31          31           0    *|
   women#prov#c.year_birth |        31           0          31     |
     women#year#year_birth |       103           0         103    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_birth_control.xls
dir : seeout
(MWFE estimator converged in 9 iterations)

HDFE Linear regression                            Number of obs   =  4,621,938
Absorbing 3 HDFE groups                           F(   1,     30) =       3.30
Statistics robust to heteroskedasticity           Prob > F        =     0.0793
                                                  R-squared       =     0.4192
                                                  Adj R-squared   =     0.4192
                                                  Within R-sq.    =     0.0005
Number of clusters (prov)    =         31         Root MSE        =     0.9161

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
          n_birth |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               1  |  -.2062963   .1135793    -1.82   0.079    -.4382562    .0256637
                  |
            _cons |   2.076518   .0461804    44.97   0.000     1.982206    2.170831
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        31          31           0    *|
   women#prov#c.year_birth |        31           0          31     |
     women#year#year_birth |       103           0         103    ?|
           women#prov#year |        92          92           0    *|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_birth_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 8 iterations)

HDFE Linear regression                            Number of obs   =  4,621,937
Absorbing 3 HDFE groups                           F(   1,     30) =       2.99
Statistics robust to heteroskedasticity           Prob > F        =     0.0942
                                                  R-squared       =     0.4183
                                                  Adj R-squared   =     0.4183
                                                  Within R-sq.    =     0.0006
Number of clusters (prov)    =         31         Root MSE        =     0.9168

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
          n_birth |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               1  |   -.207778    .120211    -1.73   0.094    -.4532816    .0377257
                  |
            _cons |   2.077121   .0488767    42.50   0.000     1.977301     2.17694
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#nsib |        13           0          13     |
                women#prov |        31          31           0    *|
   women#prov#c.year_birth |        31           0          31    ?|
     women#year#year_birth |       103           1         102     |
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_birth_control.xls
dir : seeout
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   =  4,621,938
Absorbing 2 HDFE groups                           F(   2,     30) =       1.69
Statistics robust to heteroskedasticity           Prob > F        =     0.2009
                                                  R-squared       =     0.4182
                                                  Adj R-squared   =     0.4181
                                                  Within R-sq.    =     0.0006
Number of clusters (prov)    =         31         Root MSE        =     0.9170

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
          n_birth |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               1  |  -.1998761   .1126594    -1.77   0.086    -.4299572    .0302051
                  |
women#c.sex_ratio |
               1  |  -.1772581   .1594762    -1.11   0.275     -.502952    .1484357
                  |
            _cons |   2.256026   .1857196    12.15   0.000     1.876736    2.635316
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        31          31           0    *|
   women#prov#c.year_birth |        31           0          31     |
     women#year#year_birth |       103           0         103    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_birth_control.xls
dir : seeout
(MWFE estimator converged in 7 iterations)

HDFE Linear regression                            Number of obs   =  4,621,938
Absorbing 3 HDFE groups                           F(   2,     30) =       1.82
Statistics robust to heteroskedasticity           Prob > F        =     0.1797
                                                  R-squared       =     0.4182
                                                  Adj R-squared   =     0.4182
                                                  Within R-sq.    =     0.0006
Number of clusters (prov)    =         31         Root MSE        =     0.9169

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
          n_birth |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               1  |  -.2000263   .1083159    -1.85   0.075    -.4212369    .0211844
                  |
 women#c.lngdp_pc |
               1  |  -.0226657   .1146881    -0.20   0.845    -.2568901    .2115588
                  |
            _cons |   2.208838   .6990172     3.16   0.004     .7812543    3.636421
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
         women#lngdp_pc_mi |         2           0           2     |
                women#prov |        31          31           0    *|
   women#prov#c.year_birth |        31           0          31    ?|
     women#year#year_birth |       103           1         102     |
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_birth_control.xls
dir : seeout
(MWFE estimator converged in 8 iterations)

HDFE Linear regression                            Number of obs   =  4,621,938
Absorbing 4 HDFE groups                           F(   3,     30) =       1.33
Statistics robust to heteroskedasticity           Prob > F        =     0.2846
                                                  R-squared       =     0.4183
                                                  Adj R-squared   =     0.4183
                                                  Within R-sq.    =     0.0007
Number of clusters (prov)    =         31         Root MSE        =     0.9169

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
          n_birth |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               1  |  -.1926992   .1122017    -1.72   0.096    -.4218457    .0364473
                  |
    women#c.lnsec |
               1  |   .0481612   .0329929     1.46   0.155    -.0192192    .1155415
                  |
    women#c.lnpri |
               1  |  -.0659184   .0732363    -0.90   0.375    -.2154868    .0836501
                  |
            _cons |   2.324385    .802512     2.90   0.007     .6854364    3.963333
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
            women#lnsec_mi |         2           0           2     |
            women#lnpri_mi |         2           1           1     |
                women#prov |        31          31           0    *|
   women#prov#c.year_birth |        31           0          31    ?|
     women#year#year_birth |       103           1         102    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_birth_control.xls
dir : seeout
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   =  4,621,938
Absorbing 2 HDFE groups                           F(   2,     30) =       4.71
Statistics robust to heteroskedasticity           Prob > F        =     0.0166
                                                  R-squared       =     0.4183
                                                  Adj R-squared   =     0.4183
                                                  Within R-sq.    =     0.0008
Number of clusters (prov)    =         31         Root MSE        =     0.9169

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
          n_birth |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               1  |  -.1983737   .1159113    -1.71   0.097    -.4350961    .0383487
                  |
     women#c.elig |
               1  |   .4243373   .2223988     1.91   0.066    -.0298616    .8785362
                  |
            _cons |   2.052225   .0515732    39.79   0.000     1.946899    2.157552
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
                women#prov |        31          31           0    *|
   women#prov#c.year_birth |        31           0          31     |
     women#year#year_birth |       103           0         103    ?|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_birth_control.xls
dir : seeout
(dropped 1 singleton observations)
(MWFE estimator converged in 13 iterations)

HDFE Linear regression                            Number of obs   =  4,621,937
Absorbing 7 HDFE groups                           F(   6,     30) =       4.07
Statistics robust to heteroskedasticity           Prob > F        =     0.0042
                                                  R-squared       =     0.4198
                                                  Adj R-squared   =     0.4198
                                                  Within R-sq.    =     0.0012
Number of clusters (prov)    =         31         Root MSE        =     0.9157

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
          n_birth |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               1  |  -.1884931   .1047011    -1.80   0.082    -.4023213    .0253351
                  |
 women#c.lngdp_pc |
               1  |  -.0301339   .1048314    -0.29   0.776    -.2442283    .1839604
                  |
    women#c.lnsec |
               1  |   .0650777   .0292239     2.23   0.034     .0053945    .1247609
                  |
    women#c.lnpri |
               1  |  -.0922479   .0711064    -1.30   0.204    -.2374665    .0529707
                  |
     women#c.elig |
               1  |   .5320946   .2076552     2.56   0.016     .1080061    .9561831
                  |
women#c.sex_ratio |
               1  |  -.1807607   .1207157    -1.50   0.145     -.427295    .0657737
                  |
            _cons |   2.787794   1.243998     2.24   0.033     .2472105    5.328377
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------+
               Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------------+---------------------------------------|
            women#lnsec_mi |         2           0           2     |
            women#lnpri_mi |         2           1           1     |
         women#lngdp_pc_mi |         2           2           0    ?|
                women#nsib |        13           1          12    ?|
                women#prov |        31          31           0    *|
   women#prov#c.year_birth |        31           0          31    ?|
     women#year#year_birth |       103           1         102    ?|
           women#prov#year |        92          92           0    *|
-------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_women_birth_control.xls
dir : seeout

. 
. 
. keep coef_* se_* mean_* std_*  

. drop if mi(coef_women_n_birth)
(4,949,352 observations deleted)

. gen pos = _n

. 
. save "$path3/coef_census_birth_control",replace 
(note: file /Users/Wei/Dropbox/Fertility/Results/coef_census_birth_control.dta not found)
file /Users/Wei/Dropbox/Fertility/Results/coef_census_birth_control.dta saved

.  
. *** Children outcomes ****
. use "$path2/child_gender_mort", clear 

. keep if han_p == 1 
(865,761 observations deleted)

. gl CONTROL_1 = "prov##c.year_birth year#year_birth"

. gl CONTROL_2 = "prov##c.year_birth year#year_birth prov#year"

. gen wt = 1 

. replace wt = 4 if year == 2005
(1,223,585 real changes made)

. replace die = die*100
(277,264 real changes made)

. 
. ren year yr_t

. gen year = year_birth +15

. merge m:1 prov year using "$path2/fines_economic_new", keepusing(lngdp_pc secondary primary)
(note: variable prov was byte, now float to accommodate using data's values)

    Result                           # of obs.
    -----------------------------------------
    not matched                           589
        from master                         0  (_merge==1)
        from using                        589  (_merge==2)

    matched                         9,309,439  (_merge==3)
    -----------------------------------------

. drop if _merge == 2 
(589 observations deleted)

. drop _merge 

. drop year

. ren yr_t year

. merge m:1 prov year_birth using "$path2/sex_ratio_prov_year", keepusing(male_prop sex_ratio) 

    Result                           # of obs.
    -----------------------------------------
    not matched                             0
    matched                         9,309,439  (_merge==3)
    -----------------------------------------

. drop if _merge == 2 
(0 observations deleted)

. drop _merge 

. 
. replace nbroth = 99 if mi(nbroth)
(9,215,641 real changes made)

. replace nsister = 99 if mi(nsister)
(9,215,641 real changes made)

. 
. gen lnsec = ln(secondary)
(539,353 missing values generated)

. gen lnpri = ln(primary)
(344,019 missing values generated)

. gen lnsec_mi = mi(lnsec)

. gen lnpri_mi = mi(lnpri)

. gen lngdp_pc_mi = mi(lnpri)

. 
. egen mean_lnsec = mean(lnsec)

. egen mean_lnpri = mean(lnpri)

. egen mean_lngdp = mean(lngdp_pc)

. 
. replace lnsec = mean_lnsec if mi(lnsec)
(539,353 real changes made)

. replace lnpri = mean_lnpri if mi(lnpri)
(344,019 real changes made)

. replace lngdp_pc = mean_lngdp if mi(lngdp_pc)
(42,084 real changes made)

. drop mean_lnsec  mean_lnpri mean_lngdp

. 
. gen provid = prov 

. merge m:1 provid using "$path2/Comp_Law", keepusing(law_*) nogen 

    Result                           # of obs.
    -----------------------------------------
    not matched                             1
        from master                         0  
        from using                          1  

    matched                         9,309,439  
    -----------------------------------------

. 
. cap drop elig

. gen elig = 1 if year_birth >= law_year - 6 
(9,306,676 missing values generated)

. replace elig = 0 if year_birth <= law_year - 16
(8,829,233 real changes made)

. replace elig = (16-(law_year - year_birth ))/10 if law_year - 16<year_birth & year_birth<law_ye
> ar - 6 
(477,443 real changes made)

. drop provid

. gen women = 1

. gen nsib = nbroth + nsister
(1 missing value generated)

. 
. cap erase "$path3/OCP_child_birth_control.xls"

. cap erase "$path3/OCP_child_birth_control.txt"

. foreach y_var in "die" {
  2. cap gen coef_child_`y_var' = .
  3. cap gen se_child_`y_var' = .
  4. gen mean_`y_var'  = . 
  5. gen std_`y_var'  = . 
  6. su `y_var'  [aw = wt] if han 
  7. replace mean_`y_var' = `r(mean)'  
  8. replace std_`y_var' = `r(sd)'  
  9. local j = 1 
 10. 
. foreach x_var in  "$FINE_VAR_1"{
 11. reghdfe `y_var' women#c.(`x_var') if han [aw = wt], a($CONTROL_1) cluster(prov)
 12. replace coef_child_`y_var' = _b[1.women#`x_var']   in `j'
 13. replace se_child_`y_var' = _se[1.women#`x_var']    in `j'
 14. outreg2 using "$path3/OCP_child_birth_control.xls", dec(3) append
 15. local ++j
 16. 
. reghdfe `y_var' women#c.(`x_var') if han  [aw = wt], a($CONTROL_1 women#prov#year) cluster(prov
> )
 17. replace coef_child_`y_var' = _b[1.women#`x_var']   in `j'
 18. replace se_child_`y_var' = _se[1.women#`x_var']    in `j'
 19. outreg2 using "$path3/OCP_child_birth_control.xls", dec(3) append
 20. local ++j
 21. 
. reghdfe `y_var' women#c.(`x_var') if han  [aw = wt], a(women#nsib $CONTROL_1) cluster(prov)
 22. replace coef_child_`y_var' = _b[1.women#`x_var']   in `j'
 23. replace se_child_`y_var' = _se[1.women#`x_var']    in `j'
 24. outreg2 using "$path3/OCP_child_birth_control.xls", dec(3) append
 25. local ++j
 26. 
. reghdfe `y_var' women#c.(`x_var' sex_ratio) if han  [aw = wt], a($CONTROL_1) cluster(prov)
 27. replace coef_child_`y_var' = _b[1.women#`x_var']   in `j'
 28. replace se_child_`y_var' = _se[1.women#`x_var']    in `j'
 29. outreg2 using "$path3/OCP_child_birth_control.xls", dec(3) append
 30. local ++j
 31. 
. reghdfe `y_var' women#c.(`x_var' lngdp_pc) if han  [aw = wt], a(women#lngdp_pc_mi $CONTROL_1) c
> luster(prov)
 32. replace coef_child_`y_var' = _b[1.women#`x_var']   in `j'
 33. replace se_child_`y_var' = _se[1.women#`x_var']    in `j'
 34. outreg2 using "$path3/OCP_child_birth_control.xls", dec(3) append
 35. local ++j
 36. 
. reghdfe `y_var' women#c.(`x_var' lnsec lnpri) if han  [aw = wt], a(women#lnsec_mi  women#lnpri_
> mi  $CONTROL_1) cluster(prov)
 37. replace coef_child_`y_var' = _b[1.women#`x_var']   in `j'
 38. replace se_child_`y_var' = _se[1.women#`x_var']    in `j'
 39. outreg2 using "$path3/OCP_child_birth_control.xls", dec(3) append
 40. local ++j
 41. 
. reghdfe `y_var' women#c.(`x_var' elig) if han  [aw = wt], a(  $CONTROL_1) cluster(prov)
 42. replace coef_child_`y_var' = _b[1.women#`x_var']   in `j'
 43. replace se_child_`y_var' = _se[1.women#`x_var']    in `j'
 44. outreg2 using "$path3/OCP_child_birth_control.xls", dec(3) append
 45. local ++j
 46. 
. reghdfe `y_var' women#c.(`x_var' lngdp_pc lnsec lnpri elig sex_ratio) if han [aw = wt], a(women
> #lnsec_mi  women#lnpri_mi women#lngdp_pc_mi women#nsib $CONTROL_1) cluster(prov)
 47. replace coef_child_`y_var' = _b[1.women#`x_var']   in `j'
 48. replace se_child_`y_var' = _se[1.women#`x_var']    in `j'
 49. outreg2 using "$path3/OCP_child_birth_control.xls", dec(3) append
 50. local ++j
 51. 
. reghdfe `y_var' women#c.(`x_var' lngdp_pc lnsec lnpri elig sex_ratio) if han [aw = wt], a(women
> #lnsec_mi  women#lnpri_mi women#lngdp_pc_mi women#nsib $CONTROL_1 women#prov#year) cluster(prov
> )
 52. replace coef_child_`y_var' = _b[1.women#`x_var']   in `j'
 53. replace se_child_`y_var' = _se[1.women#`x_var']    in `j'
 54. outreg2 using "$path3/OCP_child_birth_control.xls", dec(3) append
 55. local ++j
 56. }
 57. }
(9,309,440 missing values generated)
(9,309,440 missing values generated)

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
         die | 9309439    12980194    2.598598   15.90934          0        100
(9,309,440 real changes made)
(9,309,440 real changes made)
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   =  9,309,439
Absorbing 2 HDFE groups                           F(   1,     30) =       6.24
Statistics robust to heteroskedasticity           Prob > F        =     0.0182
                                                  R-squared       =     0.0225
                                                  Adj R-squared   =     0.0225
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =    15.7297

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
              die |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               1  |  -.6503059   .2603856    -2.50   0.018    -1.182084   -.1185276
                  |
            _cons |   2.782128   .0734866    37.86   0.000     2.632049    2.932208
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------+
       Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------------+---------------------------------------|
              prov |        31          31           0    *|
 prov#c.year_birth |        31           0          31     |
   year#year_birth |        94           0          94    ?|
-----------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_child_birth_control.xls
dir : seeout
(MWFE estimator converged in 9 iterations)

HDFE Linear regression                            Number of obs   =  9,309,439
Absorbing 3 HDFE groups                           F(   1,     30) =       7.54
Statistics robust to heteroskedasticity           Prob > F        =     0.0101
                                                  R-squared       =     0.0234
                                                  Adj R-squared   =     0.0234
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =    15.7220

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
              die |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               1  |  -.7836101   .2853236    -2.75   0.010    -1.366319   -.2009015
                  |
            _cons |    2.81975   .0805247    35.02   0.000     2.655296    2.984203
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------+
         Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------+---------------------------------------|
                prov |        31          31           0    *|
   prov#c.year_birth |        31           0          31     |
     year#year_birth |        94           0          94    ?|
     women#prov#year |        92          92           0    *|
-------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_child_birth_control.xls
dir : seeout
(MWFE estimator converged in 8 iterations)

HDFE Linear regression                            Number of obs   =  9,309,439
Absorbing 3 HDFE groups                           F(   1,     30) =       6.51
Statistics robust to heteroskedasticity           Prob > F        =     0.0161
                                                  R-squared       =     0.0225
                                                  Adj R-squared   =     0.0225
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =    15.7297

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
              die |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               1  |  -.6602218   .2588247    -2.55   0.016    -1.188812   -.1316312
                  |
            _cons |   2.784927   .0730461    38.13   0.000     2.635747    2.934107
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------+
       Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------------+---------------------------------------|
        women#nsib |        14           0          14     |
              prov |        31          31           0    *|
 prov#c.year_birth |        31           0          31    ?|
   year#year_birth |        94           1          93     |
-----------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_child_birth_control.xls
dir : seeout
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   =  9,309,439
Absorbing 2 HDFE groups                           F(   2,     30) =       3.29
Statistics robust to heteroskedasticity           Prob > F        =     0.0510
                                                  R-squared       =     0.0225
                                                  Adj R-squared   =     0.0225
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =    15.7297

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
              die |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               1  |  -.6548986    .264244    -2.48   0.019    -1.194557   -.1152403
                  |
women#c.sex_ratio |
               1  |   .4281579   .4306688     0.99   0.328     -.451385    1.307701
                  |
            _cons |   2.338005    .444152     5.26   0.000     1.430925    3.245084
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------+
       Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------------+---------------------------------------|
              prov |        31          31           0    *|
 prov#c.year_birth |        31           0          31     |
   year#year_birth |        94           0          94    ?|
-----------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_child_birth_control.xls
dir : seeout
(MWFE estimator converged in 7 iterations)

HDFE Linear regression                            Number of obs   =  9,309,439
Absorbing 3 HDFE groups                           F(   2,     30) =       3.95
Statistics robust to heteroskedasticity           Prob > F        =     0.0300
                                                  R-squared       =     0.0225
                                                  Adj R-squared   =     0.0225
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =    15.7296

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
              die |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               1  |   -.599917   .2480559    -2.42   0.022    -1.106515   -.0933193
                  |
 women#c.lngdp_pc |
               1  |  -.2892445   .2138371    -1.35   0.186    -.7259581     .147469
                  |
            _cons |   4.416451   1.218511     3.62   0.001     1.927918    6.904983
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------+
         Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------+---------------------------------------|
   women#lngdp_pc_mi |         2           0           2     |
                prov |        31          31           0    *|
   prov#c.year_birth |        31           0          31    ?|
     year#year_birth |        94           1          93     |
-------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_child_birth_control.xls
dir : seeout
(MWFE estimator converged in 8 iterations)

HDFE Linear regression                            Number of obs   =  9,309,439
Absorbing 4 HDFE groups                           F(   3,     30) =       3.56
Statistics robust to heteroskedasticity           Prob > F        =     0.0257
                                                  R-squared       =     0.0225
                                                  Adj R-squared   =     0.0225
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =    15.7296

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
              die |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               1  |  -.6259949   .2590049    -2.42   0.022    -1.154953   -.0970365
                  |
    women#c.lnsec |
               1  |   .1971046   .1086351     1.81   0.080    -.0247579    .4189671
                  |
    women#c.lnpri |
               1  |   .1641094   .1751866     0.94   0.356    -.1936693    .5218882
                  |
            _cons |  -1.259843   2.696602    -0.47   0.644    -6.767039    4.247354
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------+
       Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------------+---------------------------------------|
    women#lnsec_mi |         2           0           2     |
    women#lnpri_mi |         2           1           1     |
              prov |        31          31           0    *|
 prov#c.year_birth |        31           0          31    ?|
   year#year_birth |        94           1          93    ?|
-----------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_child_birth_control.xls
dir : seeout
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   =  9,309,439
Absorbing 2 HDFE groups                           F(   2,     30) =       3.98
Statistics robust to heteroskedasticity           Prob > F        =     0.0293
                                                  R-squared       =     0.0225
                                                  Adj R-squared   =     0.0225
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =    15.7297

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
              die |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               1  |  -.6559259   .2590941    -2.53   0.017    -1.185067   -.1267851
                  |
     women#c.elig |
               1  |  -.7710772   .8682761    -0.89   0.382    -2.544334    1.002179
                  |
            _cons |    2.80431   .0736244    38.09   0.000     2.653949    2.954671
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------+
       Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------------+---------------------------------------|
              prov |        31          31           0    *|
 prov#c.year_birth |        31           0          31     |
   year#year_birth |        94           0          94    ?|
-----------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_child_birth_control.xls
dir : seeout
(MWFE estimator converged in 10 iterations)

HDFE Linear regression                            Number of obs   =  9,309,439
Absorbing 6 HDFE groups                           F(   6,     30) =       2.88
Statistics robust to heteroskedasticity           Prob > F        =     0.0246
                                                  R-squared       =     0.0225
                                                  Adj R-squared   =     0.0225
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =    15.7295

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
              die |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               1  |  -.5860321    .245275    -2.39   0.023     -1.08695   -.0851138
                  |
 women#c.lngdp_pc |
               1  |  -.3425693   .2006138    -1.71   0.098    -.7522773    .0671387
                  |
    women#c.lnsec |
               1  |   .1913597   .1056662     1.81   0.080    -.0244395    .4071589
                  |
    women#c.lnpri |
               1  |   .1771503   .1722005     1.03   0.312    -.1745301    .5288306
                  |
     women#c.elig |
               1  |  -.6209336   .7314235    -0.85   0.403      -2.1147    .8728325
                  |
women#c.sex_ratio |
               1  |   .4875045   .4124396     1.18   0.246    -.3548095    1.329818
                  |
            _cons |   .0978233    2.85728     0.03   0.973    -5.737521    5.933167
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------+
         Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------+---------------------------------------|
      women#lnsec_mi |         2           0           2     |
      women#lnpri_mi |         2           1           1     |
   women#lngdp_pc_mi |         2           2           0    ?|
          women#nsib |        14           1          13    ?|
                prov |        31          31           0    *|
   prov#c.year_birth |        31           0          31    ?|
     year#year_birth |        94           1          93    ?|
-------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_child_birth_control.xls
dir : seeout
(MWFE estimator converged in 13 iterations)

HDFE Linear regression                            Number of obs   =  9,309,439
Absorbing 7 HDFE groups                           F(   6,     30) =       3.67
Statistics robust to heteroskedasticity           Prob > F        =     0.0075
                                                  R-squared       =     0.0235
                                                  Adj R-squared   =     0.0234
                                                  Within R-sq.    =     0.0000
Number of clusters (prov)    =         31         Root MSE        =    15.7218

                                       (Std. Err. adjusted for 31 clusters in prov)
-----------------------------------------------------------------------------------
                  |               Robust
              die |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
women#c.fine_6_20 |
               1  |  -.6949725   .2646784    -2.63   0.013    -1.235518   -.1544272
                  |
 women#c.lngdp_pc |
               1  |  -.3493637   .2230819    -1.57   0.128    -.8049577    .1062303
                  |
    women#c.lnsec |
               1  |   .2589772   .1223065     2.12   0.043     .0091939    .5087604
                  |
    women#c.lnpri |
               1  |     .09193   .1741476     0.53   0.601    -.2637268    .4475869
                  |
     women#c.elig |
               1  |  -.5143605   .7724786    -0.67   0.511    -2.091972    1.063251
                  |
women#c.sex_ratio |
               1  |   .3476011   .4111109     0.85   0.405    -.4919995    1.187202
                  |
            _cons |   .5975102   3.097492     0.19   0.848    -5.728413    6.923433
-----------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------+
         Absorbed FE | Categories  - Redundant  = Num. Coefs |
---------------------+---------------------------------------|
      women#lnsec_mi |         2           0           2     |
      women#lnpri_mi |         2           1           1     |
   women#lngdp_pc_mi |         2           2           0    ?|
          women#nsib |        14           1          13    ?|
                prov |        31          31           0    *|
   prov#c.year_birth |        31           0          31    ?|
     year#year_birth |        94           1          93    ?|
     women#prov#year |        92          92           0    *|
-------------------------------------------------------------+
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
(1 real change made)
(1 real change made)
/Users/Wei/Dropbox/Fertility/Results/OCP_child_birth_control.xls
dir : seeout

. egen n_birth = sum(1), by(hhid)

. 
. keep coef_* se_* mean_* std_*  

. drop if mi(coef_child_die)
(9,309,431 observations deleted)

. gen pos = _n

. 
. save "$path3/coef_child_birth_control",replace 
(note: file /Users/Wei/Dropbox/Fertility/Results/coef_child_birth_control.dta not found)
file /Users/Wei/Dropbox/Fertility/Results/coef_child_birth_control.dta saved

. 
. 
. * Robustness to adding more controls
. use "$path2/marr_policy", clear

. foreach y_var in "senior" "college" "married_ever" "late_marr" "work" "high_occ"{
  2. su `y_var' [aw = wt] if men & han
  3. sca std_men_`y_var' = `r(sd)'
  4. su `y_var' [aw = wt] if women & han
  5. sca std_women_`y_var' = `r(sd)'
  6. }

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
      senior | 5420588     7394108    .2285598   .4199051          0          1

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
      senior | 5224635     7212930    .1618858   .3683461          0          1

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
     college | 5420588     7394108    .0713284   .2573727          0          1

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
     college | 5224635     7212930    .0469342   .2114979          0          1

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
married_ever | 5420588     7394108    .9169723   .2759242          0          1

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
married_ever | 5224635     7212930    .9797343   .1409078          0          1

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
   late_marr | 5420588     7394108    .2772278   .4476299          0          1

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
   late_marr | 5224635     7212930    .1063336   .3082642          0          1

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
        work | 5420588     7394108    .9060709   .2917301          0          1

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
        work | 5224635     7212930    .7634907   .4249384          0          1

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
    high_occ | 5006602     6699586    .1432257   .3503029          0          1

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
    high_occ | 4142713     5507005    .1106718   .3137253          0          1

. sca list
std_women_high_occ =  .31372528
std_men_high_occ =  .35030292
std_women_work =  .42493845
std_men_work =   .2917301
std_women_late_marr =  .30826418
std_men_late_marr =  .44762994
std_women_married_ever =  .14090775
std_men_married_ever =  .27592416
std_women_college =  .21149794
std_men_college =  .25737266
std_women_senior =  .36834606
std_men_senior =  .41990506

. use "$path3/coef_census_ind_control", clear

. foreach y_var in "senior" "college" "married_ever" "late_marr" "work" "high_occ"{
  2. replace coef_men_`y_var' = coef_men_`y_var'/std_men_`y_var'
  3. replace se_men_`y_var' = se_men_`y_var'/std_men_`y_var'
  4. replace coef_women_`y_var' = coef_women_`y_var'/std_women_`y_var'
  5. replace se_women_`y_var' = se_women_`y_var'/std_women_`y_var'
  6. 
. gen up_men_`y_var' = coef_men_`y_var' + 1.65*se_men_`y_var'
  7. gen low_men_`y_var' = coef_men_`y_var' - 1.65*se_men_`y_var'
  8. gen up_women_`y_var' = coef_women_`y_var' + 1.65*se_women_`y_var'
  9. gen low_women_`y_var' = coef_women_`y_var' - 1.65*se_women_`y_var'
 10. }
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)
(9 real changes made)

. drop if p == 8 
(1 observation deleted)

. replace p = _n
(1 real change made)

. gen label = ""
(8 missing values generated)

. replace label = "Original" if pos == 1 
variable label was str1 now str8
(1 real change made)

. replace label = "Prov-Year FE controlled" if pos == 2
variable label was str8 now str23
(1 real change made)

. replace label = "# of siblings controlled" if pos == 3 
variable label was str23 now str24
(1 real change made)

. replace label = "Sex ratio at birth controlled" if pos == 4
variable label was str24 now str29
(1 real change made)

. replace label = "GDP /capita at age 15 controlled" if pos == 5 
variable label was str29 now str32
(1 real change made)

. replace label = "Education resources controlled" if pos == 6 
(1 real change made)

. replace label = "CSLs controlled"  if pos == 7
(1 real change made)

. replace label = "All above" if pos == 8 
(1 real change made)

. 
. 
. replace p = -p
(8 real changes made)

. foreach gender in "women"{
  2. tw (scatter p coef_`gender'_senior , xscale(alt) xline(0) xtit("Coef. on Fines at 6-20 / Std
> ", size(small))  mlabel(label) mlabp(12)  xlabel(-0.2(0.1)0.4, grid  )) ///
> (rcap up_`gender'_senior low_`gender'_senior p  , ytit(" ") horizontal lp(dash) lc(green)  lege
> nd(pos(5) ring(0) col(1) size(small) lab(1 "Senior high school") lab(2 "90% CI"))  ylabel(-9 " 
> " 0 " ") )
  3. gr export "$path4/fig_c4a.eps",replace
  4. 
. tw (scatter p coef_`gender'_college , xscale(alt) xline(0) xtit("Coef. on Fines at 6-20 / Std",
>  size(small))  mlabel(label) mlabp(12)  xlabel(-0.2(0.1)0.4, grid)) ///
> (rcap up_`gender'_college low_`gender'_college p  , ytit(" ") horizontal lp(dash) lc(green)  le
> gend(pos(5) ring(0) col(1) size(small) lab(1 "College") lab(2 "90% CI"))  ylabel(-9 " " 0 " ") 
> )
  5. gr export "$path4/fig_c4b.eps",replace
  6. 
. 
. tw (scatter p coef_`gender'_late_marr, xscale(alt) xline(0) xtit("Coef. on Fines at 6-20 / Std"
> , size(small))  mlabel(label) mlabp(12)  xlabel(-0.4(0.2)1, grid )) ///
> (rcap up_`gender'_late_marr low_`gender'_late_marr p  , ytit(" ") horizontal lp(dash) lc(green)
>   legend(pos(5) ring(0) col(1) size(small) lab(1 "Married before 25") lab(2 "90% CI"))  ylabel(
> -9 " " 0 " ") )
  7. gr export "$path4/fig_c4c.eps",replace
  8. 
. 
. tw (scatter p coef_`gender'_high_occ, xscale(alt) xline(0) xtit("Coef. on Fines at 6-20 / Std",
>  size(small))  mlabel(label) mlabp(12)  xlabel(-0.2(0.1)0.4, grid )) ///
> (rcap up_`gender'_high_occ low_`gender'_high_occ p  , ytit(" ") horizontal lp(dash) lc(green)  
> legend(pos(5) ring(0) col(1) size(small) lab(1 "White-collar") lab(2 "90% CI"))  ylabel(-9 " " 
> 0 " ") )
  9. gr export "$path4/fig_c4d.eps",replace
 10. }
(file /Users/Wei/Dropbox/Fertility/Figures/fig_c4a.eps written in EPS format)
(file /Users/Wei/Dropbox/Fertility/Figures/fig_c4b.eps written in EPS format)
(file /Users/Wei/Dropbox/Fertility/Figures/fig_c4c.eps written in EPS format)
(file /Users/Wei/Dropbox/Fertility/Figures/fig_c4d.eps written in EPS format)

. 
. 
. 
. use "$path2/marr_policy", clear

. keep if women == 1 & married_ever == 1
(5,985,227 observations deleted)

. gen mort_rate = 100 - sur_rate
(660,578 missing values generated)

. foreach y_var in "n_birth" {
  2. su `y_var' [aw = wt] if women & han
  3. sca std_women_`y_var' = `r(sd)'
  4. }

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
     n_birth | 4662619     6579862    1.999787   1.197304          0         15

. sca list
std_women_n_birth =  1.1973042
std_women_high_occ =  .31372528
std_men_high_occ =  .35030292
std_women_work =  .42493845
std_men_work =   .2917301
std_women_late_marr =  .30826418
std_men_late_marr =  .44762994
std_women_married_ever =  .14090775
std_men_married_ever =  .27592416
std_women_college =  .21149794
std_men_college =  .25737266
std_women_senior =  .36834606
std_men_senior =  .41990506

. 
. use "$path3/coef_census_birth_control", clear

. foreach y_var in "n_birth" {
  2. replace coef_women_`y_var' = coef_women_`y_var'/std_`y_var'
  3. replace se_women_`y_var' = se_women_`y_var'/std_`y_var'
  4. gen up_women_`y_var' = coef_women_`y_var' + 1.65*se_women_`y_var'
  5. gen low_women_`y_var' = coef_women_`y_var' - 1.65*se_women_`y_var'
  6. }
(8 real changes made)
(8 real changes made)

. 
. replace p = _n
(0 real changes made)

. gen label = ""
(8 missing values generated)

. replace label = "Original" if pos == 1 
variable label was str1 now str8
(1 real change made)

. replace label = "Prov-Year FE controlled" if pos == 2
variable label was str8 now str23
(1 real change made)

. replace label = "# of siblings controlled" if pos == 3 
variable label was str23 now str24
(1 real change made)

. replace label = "Sex ratio at birth controlled" if pos == 4
variable label was str24 now str29
(1 real change made)

. replace label = "GDP /capita at age 15 controlled" if pos == 5 
variable label was str29 now str32
(1 real change made)

. replace label = "Education resources controlled" if pos == 6 
(1 real change made)

. replace label = "CSLs controlled"  if pos == 7
(1 real change made)

. replace label = "All above" if pos == 8 
(1 real change made)

. 
. replace p = -p
(8 real changes made)

. foreach gender in "women" {
  2. tw (scatter p coef_`gender'_n_birth , xscale(alt) xline(0) xtit("Coef. on Fines at 6-20 / St
> d", size(small))  mlabel(label) mlabp(12)  xlabel(-0.6(0.2)0.4, grid )) ///
> (rcap up_`gender'_n_birth low_`gender'_n_birth p  , ytit(" ") horizontal lp(dash) lc(green)  le
> gend(pos(5) ring(0)  size(small) lab(1 "Number of births") lab(2 "90% CI") col(1))  ylabel(-9 "
>  " 0 " ") )
  3. gr export "$path4/fig_c4e.eps",replace
  4. 
. }
(file /Users/Wei/Dropbox/Fertility/Figures/fig_c4e.eps written in EPS format)

. 
. use "$path2/child_gender_mort", clear 

. gen wt = 1 

. replace wt = 4 if year == 2005
(1,373,603 real changes made)

. keep if han_p == 1
(865,761 observations deleted)

. replace die = die*100 
(277,264 real changes made)

. su die [aw = wt]

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
         die | 9309439    12980194    2.598598   15.90934          0        100

. foreach y_var in "die" {
  2. su `y_var' [aw = wt] if  han
  3. sca std_child_`y_var' = `r(sd)'
  4. }

    Variable |     Obs      Weight        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------------------
         die | 9309439    12980194    2.598598   15.90934          0        100

. sca list
std_child_die =  15.909339
std_women_n_birth =  1.1973042
std_women_high_occ =  .31372528
std_men_high_occ =  .35030292
std_women_work =  .42493845
std_men_work =   .2917301
std_women_late_marr =  .30826418
std_men_late_marr =  .44762994
std_women_married_ever =  .14090775
std_men_married_ever =  .27592416
std_women_college =  .21149794
std_men_college =  .25737266
std_women_senior =  .36834606
std_men_senior =  .41990506

. use "$path3/coef_child_birth_control", clear

. foreach y_var in "die"{
  2. replace coef_child_`y_var' = coef_child_`y_var'/std_child_`y_var'
  3. replace se_child_`y_var' = se_child_`y_var'/std_child_`y_var'
  4. gen up_child_`y_var' = coef_child_`y_var' + 1.65*se_child_`y_var'
  5. gen low_child_`y_var' = coef_child_`y_var' - 1.65*se_child_`y_var'
  6. }
(9 real changes made)
(9 real changes made)

. drop if p == 8 
(1 observation deleted)

. replace p = _n
(1 real change made)

. gen label = ""
(8 missing values generated)

. replace label = "Original" if pos == 1 
variable label was str1 now str8
(1 real change made)

. replace label = "Prov-Year FE controlled" if pos == 2
variable label was str8 now str23
(1 real change made)

. replace label = "# of siblings controlled" if pos == 3 
variable label was str23 now str24
(1 real change made)

. replace label = "Sex ratio at birth controlled" if pos == 4
variable label was str24 now str29
(1 real change made)

. replace label = "GDP /capita at age 15 controlled" if pos == 5 
variable label was str29 now str32
(1 real change made)

. replace label = "Education resources controlled" if pos == 6 
(1 real change made)

. replace label = "CSLs controlled"  if pos == 7
(1 real change made)

. replace label = "All above" if pos == 8 
(1 real change made)

. 
. replace p = -p
(8 real changes made)

. foreach gender in "child" {
  2. tw (scatter p coef_`gender'_die , xscale(alt) xline(0) xtit("Coef. on Fines at 6-20 / Std", 
> size(small))  mlabel(label) mlabp(12)  xlabel(-0.1(0.02)0.04, grid )) ///
> (rcap up_`gender'_die  low_`gender'_die  p  , ytit(" ") horizontal lp(dash) lc(green)  legend(p
> os(5) ring(0)  size(small) lab(1 "Child mortality") lab(2 "90% CI") col(1))  ylabel(-9 " " 0 " 
> ") )
  3. gr export "$path4/fig_c4f.eps",replace
  4. 
. }
(file /Users/Wei/Dropbox/Fertility/Figures/fig_c4f.eps written in EPS format)

.  
.  log close 
      name:  <unnamed>
       log:  /Users/Wei/Dropbox/Fertility/Results/Robust.log
  log type:  text
 closed on:  22 Apr 2020, 23:20:22
-------------------------------------------------------------------------------------------------
